Skip to main content

Full text of "Climate and the British Scene"

See other formats

■■-**: :-■■■■■ . 


and the 




Climate and the 
British Scene 


From Chaucer's sweet April showers to the 
peasoupers of Sherlock Holmes the British 
scene cannot be contemplated without 
climate entering in. In this book Professor 
Manley shows Us the best and worst of 
our much-maligned climate. He traces the 
subtle influence of sunshine and cloud, of 
dew, mist, rain, hail and snow, of heat and 
cold on the changing scene through the 
seasons. We often apologize for our cli- 
mate, but in many ways it is the best in the 
world. No great extremes of heat or cold, 
no dreaded droughts, no destructive hurri- 
canes, yet a marked seasonal rhythm with 
lots of litde surprises. The richness of plant 
and animal life, the extremely high pro- 
ductivity of our farmlands, the fleeting 
beauties of our landscape — all are closely 
linked with Britain's climate. 

It may jusdy be claimed that this is the 
first book to attempt scientifically to trace 
these intimate yet elusive relationships. 

'I have enjoyed Professor Gordon Manley's 
Climate and the British Scene as much as any 
book in the Jfew Naturalist series. What 
weather does to men and to their environ- 
ment is the theme . . . On our habits, our 
clothing, our agriculture, our food, the 
effects of the weather are considered, and 
there is a full, scholarly but lucid explana- 
tion of what causes weather, and how the 
climatic convulsions of far-past years 
affected men then and to some extent 
affect them now . . .' Country Life 

'Professor Manley's book is written for not 
too scientific people in the British Isles, 
and it fulfils its purpose admirably. The 
author is both a geographer and a meteor- 
ologist, and the particular charm of his 
book is due to the blending of his two main 
interests. Without becoming too heavy for 
the average reader, it contains a good deal 
of expert meteorological information . . . 
the book is not written for the physicist but 
for the lover of the British countryside.' 
The Times Educational Supplement 

r . 


JK *» 

P ' 





N)G \S» 















M.A. (Cantab.), D.Sc. (Manc.) 






Tbe aim of this series is to interest the general reader in the wild life of 
Britain by recapturing the inquiring spirit of the old naturalists. The 
Editors believe dial the natural pride ol die British public in the native 
fauna and flora, to which must be added concern for their coaservation, 
is best fostered by maintaining a high standard ol accuracy combined 
with clarity 01 exposition 111 presenting the results of modern scientific 
research. The plant* and animals arc described in relation 10 their 
homes and habitats and arc portrayed in the full beauty of their natural 
colours, by the latest methods of colour photography and reproduction. 



First impression 195s 
Second impression 1953 
Third impression 1955 
Fourth impression it)53 
Fifth impression 1971 

"Our Trimmer is far from Idolatry in other things, in one thing 
only he cometh near it, his Country is in some degree his Idol; he 
doth not worship the Sun, because it is not peculiar to us, it rambles 
round the world, and is less kind to us than to others; but for the 
Earth of England, tliough perhaps inferior to that of many places 
abroad, to him time is Divinity in it, and he would rather die than 
see a spire of English Grass trampled down by a Foreign Trespasser." 

George Savuae, Marquess of Halifax in The Character of 
a Trimmer, from Miscellanies, London, 1704 

"The southern artist has nothing to teach him of colour, everything 

to remind him of form for us a single landscape varies every 

minute of the day in drifting colour; to the dwellers in the lands of 
fierce sunlight it is colourless, and its form is as immutable as the 
rocks of which it is made." 

S. C. Kainf.s Smith, in Looking at Pictures, Methuen, 1921 quoted 
by Cicely M. Botley in ihc Quarterly Journal Roy. Met. S., 1931 

Printed in Great Britain 

Collins Clear-Type Press: Ixmdon and Glasgow 

All rights reserved 



Editors' Preface xiii 

Author's Preface xv 


i Introduction * 

2 The Makers of Tire Observations: Some Present-Day 

Sources of Climatolocical Data 6 

3 Some Elementary Properties of our Moist Atmosphere 24 

4 The Atmospheric Circulation over the British Isles 

and Neighbouring Seas 57 

5 Sky, Temperature and Season: Winter 93 

6 The English Spring and Early Summer 105 

7 High Summer and Autumn 119 

8 Landscape Features and their Effect on Weather- 

Part I '34 

9 Landscape Features and their Effect on Weather — 

Part II l6 3 

10 Mountains and Moorlands: The Effect of Altitude 178 

11 Snowfall and Snow-cover "97 

12 Secular Variations of Tire British Climate 213 

13 Instrumental Records: The Range of Climatic 

Behaviour 2 5 x 

14 Climate and Man 2 73 



Appendix: Climatological Data for 
Representative Stations 







i Sun, wind and cloud ; Newmarket Heath 

2 Crummock Water in June 

3 A September morning; Loch Achtriochtan, Glencoe 
4a East Midlands on a July afternoon 

4b Forest of Dean in August 

5a Low clouds over Lakeland at sunset 

5b Tailed and plumed cirrus 

6a Late summer sunset in Essex 

6b Summer sunrise in Derbyshire 

7a Tanera Beg, N.W. Scotland, in May 

7b Exmoor in March 

8a Dunstable Downs in April 

8b Allestree, Derbyshire, in August 

9 Fine January afternoon at Duddington, Northamptonshire 

10 Birchwood with hoarfrost; Little Eaton, Derbyshire 

1 1 Winter morning in a Midland city 

12 Skating on Allestree Lake, Derbyshire 

13 Ice-covered birch trees; Marley Common, Surrey 

14 May sunshine and shadows; Wood bridge, Suffolk 

15 Shropshire Plain and Wrekin in April 

1 6a Sunshine and storm over Shap Fell, Westmorland 
1 6b Evening light on River Severn, near Shrewsbury 
17 Princes Street, Edinburgh, in spring 
1 8a The Clyde estuary from above Greenock, in May 
1 8b Gorple Reservoir, Yorkshire, in early summer 
19 Sea front, Aberystwyth, in July 






20a and b Clouds over the Vale of Ock, Somerset, in July 127 

21 August evening, Studland Bay, Dorset 130 

22 Huntingdonshire fields before harvest 131 

23 Cumulus in early autumn; Beeley, Derbyshire 138 

24 Hailstorm over Buttermere, October 139 

25 Heather moor in September, Rowsley, Derbyshire 174 

26 Westerly weather in June, in the Lake District 175 

27 Cairngorms in June with snow in the corrics 190 
28a Fine September afternoon in the Lake District 191 
28b Unsettled July weather in the North Midlands 191 

29 Snow-covered landscape in bright sunshine, Hartington, 

Derbyshire 206 

30 Fine June weather in the Cairngorms 207 

31 Loch Achtriochtan, Glencoe, on an August evening 222 

32 Early summer day in Westmorland 223 

// should be noted that throughout this book Plate numbers in arabic 
figures refer to Colour Plates, while ronton numerals are used for Black- 
arid- White Plates. 



I Kew Observatory with instruments 30 

II Clouds over London at sunset 31 

Ilia Inversion fog in Somerset 46 

Illb The same view in clear weather 46 

IVa Aerial view showing summer clouds over English 

Channel 47 

IVb Aerial view of clouds over N.E. France 47 

Va March shower-clouds over Dorset coast 1 14 

Vb Isle of Man from Galloway in September 1 14 

VI Loch Maree in early April 1 15 

VII December gale along the Sussex coast 122 
Villa Rays of Aurora borealis 123 
VHIb Stormy seas breaking on the Scilly Isles 123 

IXa Daffodils in a south Scottish wood 142 

[Xb Fine weather cumulus over Baildon, Yorkshire 142 

Xa Summer morning weather over the Gareloch 143 

Xb Harvesting in County Durham 143 

XI a Cumulus cloud over the Channel, July 158 

Xlb Village cricket at Wombourne, Staffordshire 158 

Xlla Crib Goch from Snowdon, in October, showing 

distant smoke haze 159 

Xllb October day in the Outer Hebrides 159 

Xllla Colliery smoke 194 

XI I lb Rainy November day in Manchester 194 

XI Va Aerial view of cloud-sheet over Lancashire 195 

XlVb Liverpool seen through a cloud-sheet 195 



XVa Seathwaite on a fine April day 202 

XVb Ski-ing on Ben Lawers 202 

XVIa Snow on the beach, Perranporth, Cornwall 203 

XVIb Floods in Wharfedale 203 

XVII Snow in Victorian Bradford 226 

XVIIIa Norwegian glacier in retreat 227 

XVIIIb Ben Nevis snow bed in Observatory Gully 227 

XlXa Cumulus cloud over Sussex Downs 234 

XlXb Snow-laden sky in the Chil terns 234 

XX Trawlers in September gale off Norfolk coast 235 

XXIa Clouds over Irish coast, County Dublin 270 

XXIb Showery July day, Killarney 270 

XXII Last frost fair on Thames, 1814 271 

XXIIIa August afternoon on the Norfolk Broads 286 

XXI I lb Harvest in E. Sutherlandshirc 286 

XXIVa Late summer cloud-cap on Glamaig, Skye 287 

XXIVb Midsummer midnight, north coast of Scotland 287 



If a census were taken of common topics of conversation amongst 
British people it is very probable that the weather would take 
first place. So much do its vagaries enter into everyday life of the 
people that the weekend activities of millions may be directly deter- 
mined by the conditions of the moment. Twice daily at least, through 
the British Broadcasting Corporation, the Meteorological Office informs 
our fifty million people of their probable weather fate for the next 
24 hours. Whilst the emphasis would thus seem to be on the vari- 
ability of British weather and it has facetiously been said that Britain 
has no climate, only a succession of weather, in actual fact British 
climate is the most dependable in the world. November fogs, March 
winds, April showers may vary in intensity but those tremendous 
variations from year to year or fierce climatic features — the prolonged 
droughts, the delays in the 'coming of the rains', the destructive 
tornadoes, the violent cloudbursts, the destructive hailstorms, the 
abnormal snowfalls so common in other parts of the world — are 
scarcely to be understood by those accustomed to the even rhythm of 
British climate. 

Realizing the fundamental influence of climate on every aspect 
of the natural history, ecology and scenery of Britain the Editors of 
the New Naturalist early planned for a volume in the Series which 
would deal with the almost untrodden no-man's-land between the 
field of the professional meteorologist and climatologist on the one 
hand and the naturalist on the other. Just as it is a matter of extreme 
difficulty to catch the fleeting moods of the weather in the camera, 
so is it difficult to convey the subtleties of the influence of those moods 
on the life of plants, animals and man. We believe Professor Gordon 
Manley has succeeded in his difficult task as no one else could have 
done. A Past President of the Royal Meteorological Society as well as 
Buchan Prizeman and Symons Lecturer there is no doubt as to his stand- 
ing as a meteorologist. As Professor of Geography in the University 

editors' preface 

of London at Bedford College he is concerned with the application 
of these specialist studies in wider fields just as he saw their application 
in other spheres as a wartime officer of the Cambridge University Air 
Squadron. Yet we can also picture him in his little observation hut 
on the highest Pennine ridge of Crossfell, personally experiencing the 
worst of the weather conditions of which he writes, just as we can see 
him deep in his study chair seeking with glee from long forgotten 
volumes those delightful references to climate which enrich this book. 
We are confident that this volume will give as much pleasure to many 
thousands of readers as it has to us. 

The Editors 


For most of us our knowledge of the climate of the island we inhabit 
derives from our experience. Our senses register its effects not 
merely direcUy, but also through the appearance, sound and even 
smell of the world around. For some of us knowledge gained through 
the senses is reinforced by instrumental observations. It is the purpose 
of this book to try and bring together the results of qualitative appre- 
hension and quantitative recording, and to provide for those who wish 
to study any aspect of the natural history of these islands an intro- 
duction to its climatology, laying stress wherever one can upon the 
myriad ways in which the British scene in all its diversity is affected 
by the vicissitudes, regular and irregular, past and present, of our 
atmospheric environment. 

This is not a meteorological text. Little or nothing will be said of 
the upper air. I have tried to confine attention to those phenomena 
which can be observed directly from the ground by any keen walker, 
or the many other travellers about this island whose lively interest 
takes them one day into the ripening wheatfield; next day while it 
rains they see the local art gallery or sample the local cheese; a month 
later their impressions may be totally different when they return from 
work on a blustery autumn evening. Hence the properties of the sur- 
face air masses and the characteristic results of their presence at the 
several seasons are reviewed. I have also tried to provide an abund- 
ance of references for further reading by those who wish to specialise, 
especially to the many papers in the Quarterly Journal of the Royal 
Meteorological Society which has recently celebrated its centenary as the 
senior society of its kind in the world. To that journal have con- 
tributed a great number of Fellows, amateur and professional alike, 
whose desire to make some contribution to the elucidation of the 
fascinating complexities of the British climate ultimately derives from 
their upbringing in an island of such delightful variety. It is indeed 
to be hoped that a lively body of meteorologists will continue to be 


author's preface 

one of the by-products of" British weather and that at least a 
few of those who read this book will thereby be encouraged to 
embark on more serious studies. Physicists, biologists, geographers, 
geologists, mathematicians and many others can all find in the 
various aspects of meteorology a field of study of the utmost 
interest, breadth and complexity to which they may bring their 
particular training. 

I have deliberately spread a net over many byways of climato- 
logical discussion. For those who wish to embark on the straight- 
forward assessment of climatic data there are many better sources, 
some of which I have named; for those wishing to study underlying 
principles the fundamentals of meteorology are again presented in 
appropriate texts. Climate is however apprehended as a whole, and 
through several senses. Let the reader therefore try to recall not 
merely the meteorological situation, but all the feelings and associations 
of the landscapes at various seasons of which an illustration has been 

For the use of many diagrams I have to thank the Council of the 
Royal Meteorological Society, also the authors of the papers concerned, 
notably Mr. E. Gold, Mr. C. K. M. Douglas, Dr. J. Glasspoole and 
others cited in the text, who have allowed me the privilege of repro- 
duction from their publications; and the Controller of H.M. Stationery 
Office for permission to make use of officially-published daily and 
monthly weather reports and other meteorological data for purposes 
of illustration and in tables. Figures 3, 28, 29, 39, 40, 48, 49, 55, 65, 
66 and 71 are based on material, or reproduced from the Society's 
"Quarterly Journal"; Figs. 14, 27, 72 from "Weather"; the charts 
and diagrams in Figs. 4, 5, 16 to 26 inclusive, 32, 33, 35 to 38, 41, 42, 
46, 47, 58 to 61, 63 and 67 are based on H.M.S.O. publications. I 
have to thank one of the Honorary Secretaries of the Society, Professor 
P. A. Shcppard, of the Department of Meteorology at the Imperial 
College of Science, for his generous help and advice in regard to the 
first seven chapters. To Sir David Brunt and the Council of the 
Physical Society I am grateful for permission to include Fig. 74. I owe 
to Mr. Cyril Newberry a debt of gratitude for his indefatigable efforts 
to bring off one of the most difficult problems ever given to a colour 
photographer — that of catching the fugitive yet very real meteorological 
situations as they appear to the eye in typical landscapes. I am further 
indebted to Sir Nelson K. Johnson, Director of the Meteorological 

author's preface 

Office, for certain anemograms, notably Fig. 50; to Mr. Paton of the 
University of Edinburgh for the reproduction of one of his remarkable 
auroral photographs; to Dr. Balchin of Kings College in the University 
of London for the illustrations (Plate III) of inversion fog in Somerset; 
to Dr. Julian Huxley for photographs of showery weather (Plate 20) 
in the same county; to Mr. E. L. Hawke for Plate XXVIb and for 
permission to include diagrams from his papers; to Dr. C. E. P. 
Brooks for Fig. 27; to Mr. D. S. Hancock for sunshine data at Bognor 
Regis; to Mr. B. R. Goodfcllow for Plate Xlla; to Dr. P. R. Crowe 
for Fig. 71 ; to Mr. J. Wadsworth for Fig. 72. I have also to thank the 
Controller of H.M. Stationery Office for permission to use the snowfall 
and snow-cover maps (pp. 202, 204) from the Meteorological Mag- 
azine and Mr. E. F. Baxter and the University of Durham for per- 
mission to use published data from Durham Observatory. I have to 
thank Messrs. Constable for permission to quote from A Reading of 
Earth by George Meredith; Messrs. Chatto and Wind us for permission 
to quote from R. L. Stevenson's Poems; Mr. Hilaire Belloc and his 
publishers, Messrs. Duckworth, for permission to quote from The South 
Country on p. 140; and to Dr. S. Pcttcrssen and his publishers, Messrs. 
McGraw-Hill, for permission to base Figs. 6 and 1 1 on two diagrams 
from his Introduction to Meteorology. 

Other acknowledgments and references have been made wherever 
possible in the text and bibliography. If I have unwittingly made 
references to the findings of recent writers without adequate acknow- 
ledgment I have in all probability done so on account of the many 
fruitful conversations and discussions which I have been privileged to 
join at the Royal Meteorological Society; and I should like again to 
commend the readers of this book to the splendid series of papers in 
its Quarterly Journal and to the articles in its more recent monthly, 
Weather, not forgetting the abundant aspects of climatological infor- 
mation comprised in the many publications of our Meteorological 
Office in which for a short time I had the honour to serve. Tables 
and data have been largely derived from the published data in the 
Monthly Weather Report; in this I have been greatly helped by 
Mr. D. S. Brock of Westminster School, a former member of the 
Department of Geography at Cambridge, not only for the labour of 
extraction and compilation but also for the drawing of many of the 
diagrams in the text, notably Figs. 34, 52, 69 and 70, and the necessary 
abridgment of many weather maps for small scale reproduction. I 
C.B.S. rvii b 


have again to thank Sir N. K. Johnson, Director of the Meteorological 
Office, Dr. J. Glasspoole of the Climatological Division and the many 
members of the staff who have helped me for the opportunities so 
generously provided ; not least for the use of tabuladons already made, 
in the task of bringing such things as rainfall averages up-to-date. 
In compiling such tables I have been at pains to choose a number of 
localities of rcprescntadve interest to supplement those for which 
material is already available in the well-known standard text (E. G. 
Bilham, The Climate of the British Isles, London, Macmillan, 1938). 
Lastly no one who looks into the origins of our climatological informadon 
can fail to acknowledge an indmate debt to the long line of patient 
observers who have maintained a great tradidon since the days of 
Robert Hooke; a part indeed of that observadon of nature whose deep 
appeal we can ever share. 

I have to thank Miss Alison Birch for much assistance with the 
reproduction of drawings and maps. I am grateful to Miss C. M. 
Bodcy for drawing my attention to the quotation from Mr. S. C. Kaincs 
Smith's Looking at Pictures, 1921; and to Messrs. Methuen, the pub- 
lishers, for allowing me to quote it here. Lastly, I have to thank 
Mrs. Audrey Hitchcock, a former student of the Department of 
Geography at Bedford College, for completing the index as well as 
for assistance in bringing the tabulations in the Appendix down to 
1949. Tabulations are throughout based on Meteorological Office 
published data, wherever available; the reductions, calculations and 
averaging I have largely performed myself. References are in greater 
profusion than is usual in a book of this type; but the profusion of 
by-ways in meteorology is now so great that I hope it will thereby be 

found more useful. 

Gordon Manley 

Note to the Fourth Impression 

In this, the fourth impression, the opportunity has been taken to 
make minor additions, and to incorporate a number of corrections to 
which several correspondents have been kind enough to draw my 
attention; I am glad to be able to thank them. Certain tables and 
diagrams have been brought up to date (1961); I have to thank Miss 
Elizabeth M. Shaw for assistance with Fig. 65. With regard to the 
tables in the appendix it should be added that the Meteorological 
Olfice has now published averages for 1921-50. p 

chapter 1 


Calm was the day, and through the trembling air 
Sweel-breat/iing %fi$hyrw did softly play a gentle spirit 


Climate may be defined as an expression of our integrated experi- 
ences of "weather". As a word it comes to us from the Greek, 
from long before the days of instruments, so we may well look into the 
classical background. In the countries of the Mediterranean whence 
we derive many of the concepts implicit in our speech the over- 
whelming power of the summer sun has always impressed mankind. 
The short, sharp shadows of July, the white glare, the very rapid 
evaporation and the scorched remnants of grass are today a vivid 
recollection in the minds of many travellers. Throughout the Medi- 
terranean coasdands there is moreover a very high proportion of 
sunny hours in the summer months. It is no wonder, therefore, that 
the Greek philosophers considered that climate was primarily depend- 
ent on the altitude of the sun, that is on latitude. Greek geographers 
were accustomed to recognise seven "climates" between the Equator 
and the Poles. 

Around the Mediterranean and throughout much of South- 
Central Europe the seasonal rhythm is simple and relatively well- 
defined. The cloud, wind and intermittent rain of winter along the 
Mediterranean coasts stand in sharp contrast with the drought of 
summer. Further inland to the northward, there is a broad region 
where a definite season can be expected during which frost occurs, 
and there is a risk of snow; while through Burgundy and South-Central 
Europe the lowland summer is always warm enough for the ripening 
of crops. Occasional catastrophic thunderstorms and floods may 
occur causing local devastation but even the coolest and most cloudy 


summer is not calamitous over a wide area in the sense that is found 
towards the north-west margins of Europe. 

Gradually as men move towards the north-west coasdands of 
Europe the wind plays an increasing part in their consciousness. For 
many months it can be said that the sun gives light, and at times an 
agreeable warmth. But it is the quality or "feci" of the air that we 
in the British Isles perceive first as we go out of doors. Ultimately 
this arises from the fact that the air arrives from a variety of different 
sources, and undergoes a varying degree of modification on its way 
towards us. The climate of the British Isles and die consequent 
appearance of the landscape owes much to latitude, notably in respect 
of the great seasonal variation in the amount and intensity of light; 
but still more is due to our position and maritime surroundings, 
especially with regard to wind. 

Not that the Greeks were unmindful of the effects of wind. Winter 
in the Mediterranean is a season during which, as depressions pass, 
markedly different types of air pervade those coastal chaffering-places 
beloved by Mediterranean man. Boreas was the north wind, coming 
off the snowy continent of Europe to give cold clear air with good 
visibility and a nearly cloudless sky over the country round Athens; 
Notos, from the south, was recognised as warm, humid and enervating 
by comparison. In Rome blustery tramontana and languid sirocco 
play the like part. But the Atlantic shores of Europe offer the same 
contrasts with greater boisterousness throughout the year. It is a 
northern composer — Sibelius of Finland — who has given expression 
in music to the tremendous majesty, persistence and interminable 
energy of the northern winter storms. The northern mythology of the 
early Scandinavians characterised die sun as feminine. We can 
perceive the hint in the actions of those descendants who through the 
centuries have thrust, crept or clawed their way into the maternal 
bosom of Southern Europe, by contrast with those who faced the open 
Adantic and even ventured across it. 

The climate of the British Isles is such that the inhabitants enjoy, 
but arc not subordinate to, the power of the sun. It has accordingly 
been stigmatised by Latin Europe. Tacitus left a renowned note on 

Plate i 

Newmarket Heath, Suffolk; in spring sunshine, May. North Sea, strato-cumulus 
in moderate N.E. wind, temperature 55 . 


PLATt 2 


the subject: "The climate is objectionable with frequent rains and 
mists, but there is no extreme cold". Dumas gave vent to the views 
of the romantic Sturm-and-Drang period, "L'Angleterre est un pays 
ou le solcil rassemble a la lune". By contrast, more discriminating 
observers have often found room for praise; and Englishmen them- 
selves, especially those who have not dwelt for long elsewhere, or have 
resisted the seductions of lands nearer the equator, are evidendy very 
proud of their climate. As a Venetian Ambassador said in 1497, "The 
English are great lovers of themselves and of everything belonging to 
them". Of Charles II, it is recorded that he never said a foolish thing 
and never did a wise one; a view which would not be entirely upheld 
by the Royal Society, formed with his encouragement in 1663. Charles 
was a keen observer; and his saying that "The English climate is the 
best in the world. A man can enjoy outdoor exercise on all but five 
days in a year" has been echoed by most energetic Englishmen. 

In 1944 on a delightful May morning in Suffolk, an American 
enlisted man said, "I like this weather of yours. You can work all the 
year round without sweating". From America, too, the late Professor 
Ellsworth Huntington acknowledged the advantages of the climate 
of south-east England, no fight tribute from such an energetic scion 
of Yale. 

Appreciation of the British climate depends largely on temper- 
ament. That it has not been conducive to idleness has been reflected 
in the characteristics of the people; be it remembered that the urge 
to go and do something useful, to keep moving, to use one's intelligence, 
to protest against indolence, to stir up controversy and to censure the 
offcomer is also increasingly marked to the northward. Unreasonable 
activity and exertion are, however, gently damped down — the English- 
man's own expression. Undue assertiveness in colour, music, archi- 
tecture, opinion or sentiment is out of keeping; it is "not done"; 
gentle gradations of colour and fluctuations of mood are associated 
with the lack of the sharp shadows, the harder lines and fiercer con- 
trasts of more southern lands. Some, be it noted would say "cruder 
contrasts" ; others derive from the sharpness of contrast a stimulus to 
activity. Accordingly opinion as to the goodness or badness of the 

Plate a 

Orummock Water, Cumberland: breezy afternoon, June. Fair weather cumulus 
ai about 4,000 I'ect with fresh N.E. wind, temperature 6o°. 


British climate is likely to vary immensely from time to time between 
different leaders of opinion in this and other countries. Judgment can 
only be given by results, particularly in the realm of stockbreeding 
according to Mary Borden; and we may extend the argument to 
the results in the realm of originality of mind and accomplishment 
among the inhabitants for at least fifteen centuries past. To what 
degree this is the result of racial mixture, natural selection, or environ- 
mental influence it is not the purpose of this book to explore. We shall 
instead embark upon the discussion of climate, having especial regard 
to its significance as a factor in the moulding not merely of landscape 
but of the many other elements that go to make up the British scene. 
Shakespeare's intuitive appreciation of the qualities of his ideal 
island led him to put into the mouth of Caliban — the unlettered but 
sentient native — 

Be not afraid, the isle is full of noises, 

Sounds, and sweet airs, that give delight and hurt not. 

Yet there are but few days when this cannot equally be said of 
that English countryside which Shakespeare knew. That our airs 
give delight and hurt not goes far to explain the attractions as well as 
some disadvantages of this country of ours; a point of view to be 
developed in later chapters. 

The many-faceted British scene indeed has the fascination of a 
well-cut diamond by contrast with the crude regularity of a simple 
crystal. Like a diamond it owes much to the artifice of man; the 
eighteenth-century landscape gardeners are responsible for the char- 
acteristics of much of Southern and Central England. Eighteenth- 
century landowners were likewise active in Scotland as the varied 
charms of Roxburgh and the Lothians still show. But the glitter of the 
facets owes much to the fight; in regard to our varied British landscape 
this is primarily a result of our climate. We shall try to embark on a 
discussion of British weather particularly as its complex effects are 
perceived by our several senses when we go out-of-doors. Even 
indoors; our satisfaction with what many Americans arc apt to cherish 
as the ruefully enjoyable ineffectiveness of our ancestral domestic 
heating arrangements also owes much to the curious qualities of our 
outdoor climate. In humid air approaching saturation when there is 
little or no wind vigorous exertion quickly leads to discomfort through 
overheating, although with the same amount of clothing it is too cold 


to sit still. Our normal winter indoor clothing allows for this com- 
promise at somewhere between 50 and 55° in damp weather. With 
a bright source of radiant heat such as a fire, we sit and work quite 
comfortably in rooms between 55 and 6o°, a temperature at which 
the gentle evaporation from the skin in the humid air does not add to 
that sense of dryness to which our skins are unaccustomed. Fires are 
generally dispensed with, in the daytime, as soon as the noon temper- 
ature begins to exceed 6o°, that is from early May to early October in 
Southern England. But it is significant that Continental ideas of 
heating are beginning to spread in our greatest mart for fashion, 
propaganda and advertisement, and the urbanised monotony of the 
great shops and hotels of London might be exchanged without com- 
ment with that of like institutions in many great cities abroad. Is it 
any wonder that our city populace is showing signs of forgetting that 
the fundamental needs of life — food, water, shelter and transport — 
arc still subject to one of the most erratic climates in the world? 

Weather Maps 

For those unfamiliar with the notation used in weather maps, of which a 
number appear in subsequent chapters, a short key will serve; Fig. 4 on p. 31 
may be referred to. The winding lines are isobars; by joining places with equal 
barometer readings (after reduction to sea level) they illustrate the distribution and 
shape of regions of low and high atmospheric pressure at any given time. Pressures 
are in millibars (1000 mb. can be taken as 29.53 "inches of mercury"); mean 
annual pressure at Kew is 1015 mb. Figures beside stations arc temperature in 
degrees F. Wind direction is shown by the shaft of the arrow, force by the number 
of fleches. One short and two long fleches means force 5 on the Beaufort scale, i.e. 
between 17-24 m.p.h. or, in sailors' language, a fresh breeze. Four long fleches 
means force 8 (38-45 m.p.h.) and is described as a gale. Lack of an arrow means 
a calm. State of the sky and prevailing weather are indicated by shading of the 
circle or by a symbol at the head of the arrow. ThusO = blue sky, not more than 
one quarter clouded; (Q (fj) (2D increasing proportions of cloud, <JTJJ> overcast; • rain, 
* snow, * sleet; = fog, CO haze. Fronts, that is lines on cither side of which 
there is a marked difference in the qualities of the air, arc shown :^».*.x warm 
front, at which warmer air is advancing and rising over cooler air ahead; the 
symbol indicates direction of advance of the system. 

4AA. a cold front, i.e. cold air advancing and undercutting wanner air ahead; 
symbols again show the direction in which the front is advancing. 
■*■»*«, an occlusion, across which there is in general little difference in temperature 
at the ground; the difference lies above and at some altitude warm air is being 
elevated, giving rise to extensive cloud and often rain or other precipitation. 



Look how the gusty sea of mist is breaking 
In crimson foam, even at our feet! it rises 
As Ocean at the enchantment of the moon 
Round foodless men wrecked on some oozy isle. 
Shelley: Prometheus Unbound 

A ppreciation of the climate of the British Isles was perforce almost 
l\ entirely qualitative until with the invention and diffusion of 
measuring instruments the accompanying spirit of scientific inquiry 
became widespread in the seventeenth century. For centuries before 
that, however, we can be sure that the sailors along the North Sea 
coasts had a very shrewd idea of the behaviour of the air, and that the 
farmers inland acquired a fund of sound knowledge regarding the 
merits of their fields under varying weather conditions. The sites of 
Anglian, Saxon and Danish villages, of Celtic croft and Norse farm- 
stead betoken a keen eye for climatic advantage. Such knowledge, 
however, played little part in die minds of the literate; rarely there 
were signs of grace, notably in the Oxford clergyman who from 
1337-44 kept a daily record of the direction of wind and the occur- 
rence of rain. From Elizabeth's day onward a "wind book" appears 
to have been kept at the Admiralty; and the Elizabethan predecessors 
of to-day's geographers endeavoured to give some hint of the climatic 
features of the counties of England. Of Durham Speed (1610) wrote, 
"The air is subtle and piercing, and would be more, were it not that 
the vapours of the North Sea do much to dissolve her ice and snow." 
In such a way the Elizabethans endeavoured to explain the fact that 
the mean annual frequency of mornings with snow-cover is but eight 
at Tynemouth, eighteen at Durham, and eighty in those parts of 
upper Teesdale lying on what they knew as "Fiends Fell" — that bleak 


Crossfcll region which above 2,000 feet has been forsaken of man's 
habitations since at least 500 B.C. and perhaps since time began. 

Instrumental recording began here and there soon after the inven- 
tion of the thermometer and barometer. Early use of both instruments 
was made by Robert Hooke; his MS. for 1664 is preserved at the Royal 
Society. John Locke kept instrumental records as early as 1667. 
Instruments were, however, very imperfect, and the necessity of making 
adjustments and corrections was not fully understood, so that we can 
as yet make little use of our observations of temperature and pressure 
until after 1750. 

Rainfall was first recorded over a considerable period in England 
by a most interesting character in a most appropriate place. A Lan- 
cashire squire, Richard Towneley, living at Towneley Hall near 
Burnley on the flanks of the Pennines, designed his own rain-gauge 
with much ingenuity and kept a record from 1677 to 1704. Let it 
not be forgotten that Towneley in this respect was a pioneer; far away 
from opportunities of discussion with the men of the nascent Royal 
Society, he showed that spirit of inquiry and desire to keep an accurate 
record which has appeared in one form or another in thousands of his 
countrymen. The same spirit of independent inquiry combined with 
a desire for accurate recording and transmission of knowledge flowered 
earlier in Wycliffe and Coverdale, and, some would add in Bede; it 
appears in Christopher Saxton, James Cook and William Scoresby; 
in John Dalton, Adam Sedgwick and John Phillips. The results of 
observation were intelligently applied by George Stephenson, Edward 
Pease, Joseph Whitworth, to whom many descended from the dales 
can be added today. Here we have named but a few of the men who 
were bred of the stocks which, like the sheep and cattle, have flourished 
for so long along the flanks of the windy northern moorlands; such 
men have also been part of the British scene. Elsewhere too our hilly 
districts appear to have been the original home of much of the 
observational science on which our modern technical advances rest: 
this seems to be as true of meteorology as of other sciences. Tyndale, 
Norden, "Strata Smith" and Huxley derive from Somerset. East 
Anglia, the home of so many early naturalists and landscape painters, 
has also provided its quota of early observers of weather; one of our 
earliest instrumental records (1673-4) comes from Wrentham near 
Lowestoft. From Norfolk the Marsham phenological records began 
in 1736. 




Nevertheless even a moderate degree of instrumental accuracy 
was not attained in a day, and still there is room for improvement. 
The eighteenth century saw considerable progress in the realm of 
meteorology; scattered here and there about England and Scotland 
many men began to use the slowly improving types of instruments and 
to add daily readings to their "weather diaries" already in fashion. 
Thomas Barker, the brother-in-law of Gilbert White the naturalist, 
kept meteorological records at Lyndon in Rudand from 1736 to 1798. 
John Huxham, a physician at Plymouth, was another early observer 
of note; indeed during the 18th century the contribution of the medical 
men to our climatological knowledge was greater than that of any 
other group. 

Widi his northern contemporary the Reverend Thomas Robinson 
of Ousby beneath Crossfcll, William Derham, whose rainfall record 
at Upminster in Essex covers the years 1697 to 17 16, was one of the 
first clergymen to keep instrumental records, and thus to set a fashion 
for many of his colleagues in later days. Derham also observed an 
early type of thermometer, but under conditions which are not easy 
to interpret. Robinson's readings have not come down to us; but he 
left us the first account of the helm wind (1696). 

An unknown Edinburgh medical man kept the first Scottish 
temperature observations ( 1 73 1 -36) that have been reduced to modern 
standards. To satisfy the curiosity of the reader before going farther, 
it may be added that none of those early rainfall, temperature, or 
wind observations indicates a range of climatic variation greater than 
might be expected at the present day; and, of course, this is borne 
out by the seventeenth-century botanists who recorded the date 
of flowering of various plants, of leafing of trees, and the date of 

Slowly it became possible to put these observations together: a 
considerable assemblage of rainfall data was made by John Dalton 
in 1799. Dalton rivals Barker as one of the most determined meteoro- 
logical observers who ever lived; his daily records ran from 1787 to 
his death in 1844. He began recording rainfall at Kendal in 1787, 
with the result that Kendal soon took rank as the wettest place known. 
However, it was not until 1840 that a sufficient number of rainfall 
records became available to enable Joseph Atkinson, another amateur 
meteorologist from Carlisle, to compile the world's first tentative 
rainfall map. 


The maintenance and compilation of meteorological observations 
demands considerable assiduity, as men were quick to note. Ralph 
Thoresby of Leeds visited Towneley in 1697 and saw the rain gauge; 
in his journal he notes that he thought much of keeping a similar 
record, but decided that he was not sufficiently patient. Dr. Short 
of Sheffield in 1750 was more downright; "for it being a dry subject, 
most gentlemen are soon weary of it," he wrote. In other countries 
support for meteorological recording 
was not uncommonly forthcoming long 
before 1800 at State observatories 
and the like. In Britain, almost all 
our earlier records were kept by inde- 
pendent amateurs. In this respect 
meteorology resembles all the other 
observational sciences. Not the least 
of the attributes of this island of 
ours is that if the normally cultivated 
countryside is left to itself, climatic 
factors lead to the establishment of a 
wild vegetation of great variety, on 

which order can only be imposed by prolonged co-operative effort. 
Without doubt the analogy holds with regard to our scientific accom- 
plishment as a people. 

Co-operative effort was first essayed through the establishment of 
our many scientific societies. Among these die Royal Meteorological 
Society, founded in 1850, is the senior body of its kind in the world. 
It belongs to that Chartist decade in which the Chemical Society and 
the Rochdale Pioneers began their operations; a decade in which 
public health, water supplies, sanitation and engineering were also 
demanding co-operative effort, while at the same time leading many 
men to keep more precise records of weather. The Repeal of the Corn 
Laws followed the miserable summer of 1845. 

Before the days of official organisations, temperature, pressure and 
other elements were observed in various ways For a long time, shade 
temperatures were thought to be sufficiently comparable if taken on a 
north-facing wall. Indeed, during the eighteenth century diere was a 
long period during which many men kept their records by reading a 
thermometer inside a room without a fire. Such records are of course 
of little value unless very carefully used. This curious fashion probably 

F10. 1 

Bilham's modification of the 

Stevenson screen 



arose among some of the doctors. In 1723, Jurin, then secretary of the 
Royal Society, recommended this method, perhaps because it suited 
his own arrangements in London, but more probably because as a 
medical man he knew that many men in cities pass most of their time 
indoors. Later in the century, however, the growing interest in gar- 
dening undoubtedly led to the keeping of better temperature records, 
here and there about the country. James Six, a gardener, invented the 
familiar "Six's maximum and minimum" in 1782. More reliable 
minimum thermometers began to come into use about 18 10, but were 
not generally used until the forties. Before that date most observers 
recorded "fixed hour" readings twice or thrice daily. Gilbert White 
himself took observations of temperature, pressure and rainfall from 
1768 to 1793; and scattered through the country were not only the 
doctors but men such as Holt of Liverpool with his keen observations 
on Lancashire agriculture (1794). Later we find that the Royal 
Horticultural Society's record in London was begun in 1825. Many 
early records were kept by the London instrument makers and we may 
conclude that they were not forgetful of the advertisement thus given. 
Among the best of these is John Cary's record at his shop in the Strand 
(1786- 1 846). The present-day record kept by Negretti and Zambra 
in Regent Street continues the honourable tradition begun by men such 
as Ayscough of Ludgate Hill in the 1750's; other examples include 
Casartelli of Manchester and Pain of Cambridge. 

A long time elapsed before it was fully realised that the air temper- 
ature is not necessarily given by a thermometer on a north wall. In 
such circumstances, a thermometer, for example, on a warm day in 
March may very well be reading much too low, as it is recording the 
temperature of the wall rather than that of the air. Hutchinson, the 
remarkably lively Liverpool harbour master, tried to solve tins problem 
by keeping his thermometer (1777-93) under a table on his roof. 

In the nineteenth century increasing attention was given to these 
problems and from 1823 we learn of small groups founding "Meteor- 
ological Societies" for their discussion. From 1847 onward official 
encouragement was given to the collection of records kept under 
standard conditions. These were summarised by Glaishcr, following 

Plate 3 

Loch AcjrnuocirrAN, Glencoc, Argyll: September morning. Low stratus and 
fracto-stratus among mountains after drizzle, temperature 55 . 


Cyril Stwbtrry 

Julian lluxlrr 



his appointment (1840) as Superintendent of Meteorological Obser- 
vations at Greenwich, in the Registrar-General's Quarterly Returns. 
After 1866 the Stevenson screen gradually came into use, and slowly 
began to supersede older methods of exposure; it was invented by 
Thomas Stevenson, the Scottish lighthouse engineer, who was also the 
lather of Robert Louis Stevenson. Much encouragement was given 
to its use by the Royal Meteorological Society, already founded in 
1 850, which organised in die sevendes a scries of stadons all over the 
country using this uncxcepdonable screen. The Stevenson screen is, 
of course, simply a box with ventilated sides in the form of louvres, 
and a ventilated floor and top; above the top is an air-space and a 
second roof. Such an arrangement ensures that the thermometers 
inside the box are recording the temperature of the air. The purpose 
of the top and sides is to prevent direct radiation from the sun reaching 
the thermometers; the slats forming the floor are a protection against 
radiation from the ground and surrounding objects, which may be 
very considerable, especially in hot dry weather. Anyone can perceive 
this who walks near a soudi-facing brick wall about sunset after a fine 
summer day. 

Official recognition of the value of a network of standardised 
meteorological observations evolved, as we have seen, somewhat 
tardily in Britain, although advocates were not wanting. Even in the 
seventeenth century Hooke recognised that comparable data from a 
network of stations would be valuable. Another fillip to the exchange 
of records was given by Turin. From 1773 onward observatioas of 
some kind were maintained at the King's private observatory at Kew, 
though not very regularly; and in 1774 the Royal Society initiated 
regular daily readings in London. Oddly enough, although Greenwich 
is the oldest surviving observatory in the world (1697) standard 
meteorological observations were only begun under Glaisher in 1841. 
With his encouragement the world's first synoptic weather maps were 
compiled with die aid of the newly-invented electric telegraph and for 
some time were shown daily at the Great Exhibition in 185 1. The stage 

Plate 4 

a. East Midlands: warm afternoon verging on unsettled, July. Well-developed 

later afternoon cumulus. Light westerly wind, afternoon maximum 74 . 

b. Forest op Dean, Gloucestershire: August afternoon. Moderate S.W. wind, 

rather humid; 68°, heavy strato-cumulus and cumulus. 



was set for an official organisation on a larger scale; and the Meteor- 
ological Office was founded as a department of the Board of Trade in 
1854. For some years however, the issue of forecasts was frowned 
upon, while other countries went ahead. Daily synoptic charts and 
the issue of forecasts were finally adopted in the seventies, together 
with the collection of strictly comparable climatological data. 

The decades 1840- 1890 were the great days of the amateur. 
Around the Royal Meteorological Society there gathered a body of 
active observers. Among them were many clergymen; one of their 
best representatives was the Reverend Leonard Jenyns, vicar of 
Swaffham Bulbeck near Cambridge from 1823-53, wno published 
his Observations in Meteorology in 1858; and a very pleasant work it is, 
not without profit to anyone who chooses to read it to-day as an 
epitome of what can be done by a keen amateur observer. As early as 
1863 we can find the contributor of a paper to the Quarterly Journal 
surmising that the conflict of 'equatorial' and 'polar' streams of air 
was a necessary concomitant of what we now call 'depressions'. That 
this amateur tradition has continued to thrive is shown by the long 
series of observations kept by such men as Thomas Backhouse of 
Sunderland (from 1857-1915), John Hunter of Belper (from 1877- 
1931) and John Dover of Totland Bay (from 1886- 1948), one of whose 
diagrams is reproduced on page 90. Among other observers the 
phenomenal energy of Clement Wragge who for many months 
climbed Ben Nevis daily (1880-81) can fitly be recalled. More recently 
the extensive kite-flying experiments of Mr. C. J. P. Cave of Stonor 
Hill did much to add to our knowledge of the behaviour of the atmos- 
phere before aircraft came into use. One of Mr. Cave's superb cloud 
and landscape photographs will be found as PI. Va, opposite p. 1 14. 
It is indeed a pleasure to be able to acknowledge permission to include 
the work of die late doyen of the amateur meteorologists of this 
country, some of whose photographs have been published in book 
form by the Cambridge University Press. 

Our network of rainfall recording stations received a great fillip 
through the establishment, by George Symons in i860, of the British 
Rainfall Organisation. Gradually the number of observers, largely 
voluntary, rose to over 5,000. In 1901 the direcdon was taken over 
by Dr. Hugh Robert Mill, one of the most eminent of British geo- 
graphers and meteorologists; under his administradon it became 
possible to map the distribution and rainfall with remarkable accuracy 



for almost any part of Great Britain. No better testimony to the great 
"amateur tradition" of British scientific inquiry can be found than this 
noticeably successful enterprise for the collection and dissemination 
of rainfall statistics. These, of course, have proved invaluable for 
numerous practical purposes, such as the design of reservoirs. Since 
Dr. Mill's retirement in 191 9 the administration has been conducted 
from the Meteorological Office; and the carefully-compiled annual 
volumes of British Rainfall under Dr. Glasspoole's editorship are well 
known as a source of detailed information. 

The story of the early investigations of rainfall is interesting and 
mention should be made of the efforts of the pioneer amateurs to 
record rainfall in more remote places. Especially is this true of our 
mountain districts. Nowadays it is common knowledge that places 
can be found here and there among our highest mountains with 
exceptionally heavy rainfalls. But in the eighteenth century the 
greater rainfall to be expected in hilly districts was scarcely even 

In 1 769, Heberden, a London physician and a member of a family 
of which several generations took an active interest in meteorology, 
showed that if a raingauge were placed at some height above the 
ground, e.g. on top of a tower, less rain was normally caught than by a 
gauge at ground level. For this purpose he used the towers of West- 
minster Abbey. This work drew the attention of John Gough, the 
famous blind "natural philosopher" of Kendal, who also encouraged 
the young John Dalton to make instrumental observations. Between 
1 787 and 1790 Gough set up a number of gauges at places near Kendal 
and it so happened that those placed on higher ground recorded in 
general less rain. There is little doubt that the reason for this lay in 
the imperfections of exposure. If a raingauge is placed in a very windy 
situation it is invariably found that less rain is caught by comparison 
with a more sheltered position nearby. This is due to the setting up of 
eddies round the gauge itself in windy weather, which have the effect 
of carrying a proportion of the falling raindrops outside the perimeter 
of the gauge. Obviously the same factors operate when the gauge is 
exposed on a high roof or tower, unless the roof is exceptionally broad 
and flat. 

But many years passed during which the opinion prevailed that 
there should be less rain at a higher level. Theoretical reasons for 
this were adduced. Hutton, the Scottish geologist, had laid down a 



theory of rainfall in 1784, according to which rain resulted from the 
mixing of air currents of different temperatures. He based this argu- 
ment on the known fact, recendy determined by the eighteenth 
century physicists, that the capacity of air for water vapour increases 
more rapidly than the temperature. To illustrate: Saturated air at 
48°F. contains 87 grams of water vapour per cubic metre. At 40 
the weight will be 6-5 grams; at 56 11-5. If, therefore, we mix one 
cubic metre of saturated air at the lower temperature (40 ) with one 
at the higher temperature (56 ), we shall have two cubic metres at 
the temperature intermediate (48 ); but we shall also have rather 
more water vapour ( 1 8 grams) than the two cubic metres of air can 
themselves sustain (17-4 grams) at that intermediate temperature. 
Hence some of the moisture must be condensed whenever two satur- 
ated masses of air differing in temperature are mixed. 

14 1 

40 50 

A, B : Air saturated C: Air suyw r-satu rated ; some moisture, must condense 

Fio. 2 

Now in nature this process does repeatedly occur and indeed it is 
extremely important; but the condensation so provoked is in general 
observed as a narrow belt of fog or low cloud along the boundary 
between the two air masses concerned. We may expect some mixing 
to take place along such a boundary; but the quantities of moisture 
thus condensed are very small and remain suspended in the air in the 
minute droplets of cloud or fog. Rarely if ever is diere sufficient 



provocation in those circumstances for them to aggregate into the size- 
able drops whose rate of fall justifies their description as drizzle, let alone 
rain. It may be mentioned here that a round figure for die diameter 
of cloud droplets is r^ss of an inch: drizzle droplets, which drift rather 
than fall, are of the order of 53 of an inch; raindrops are of the order 
of 5 to ; of an inch. Adjacent to the ground the stratification of the 
nearly saturated lowest layers of the atmosphere in the cooler months 
leads to the extensive development of fog along the boundary between 
them; as we shall see, this process must be invoked with regard to the 
thickening of nocturnal radiation fogs. 

But as regards rain, it is evident that following such a doctrine, 
the cooler the air the less would be the resulting rainfall; hence early 
meteorologists were much puzzled by the fact that winter rainfall 
tends in die north to exceed that of summer. It was not for many 
years that it became widely recognised that such mixing processes 
might explain thin cloud and fog, and possibly a little drizzle; but were 
not adequate to explain rainfall. Moreover all normal cloud and fog 
is far too thick and contains far more condensed moisture than could 
be produced by mere mixing. 

In the early nineteenth century, Dalton and others began to recog- 
nise that if a mass of air is for one reason or another compelled to rise 
and to expand, its temperature must be lowered; and it was Dalton 
who, in 1 799, explained the behaviour of water vapour in the atmos- 
phere. But considerable time elapsed before die necessary consequences 
of this cooling of air by expansion were worked out. Dalton himself, 
faithful as far as one can judge, to the memory of his early friend and 
teacher (Gough of Kendal) continued to skirt round the question 
whether rainfall did or did not increase on mountains. Another 
view, not uncommonly held, was that the moist air from the Atlantic 
impinged against die mountain sides and was there cooled by mixing 
with "the cool air adjacent to the mountain slopes". When men 
began to measure surface temperatures of the air on mountains this 
view again became untenable. 

Conviction, however, of the existence of excessive mountain rain- 
falls again arose through the efforts of amateurs. In 1836 a certain 
Mr. Beck, a newly-arrived resident at Esthwaite near Windermere, 
recorded his rainfall. It was so much in excess of that at Kendal 
as to rouse the attention of others, notably John Fletcher Miller of 
Whitehaven. He set out on a very thorough investigation; first he 
c.b.s. G 



supplemented the observation in his own garden by erecting a rain- 
gauge on the adjacent church tower. In the autumn of 1844 he went 
further and set up the first rain-gauge at Seathwaite, the group of 
farmsteads at the head of Borrowdalc, which has since become known 
to thousands of Victorian and later schoolchildren as "the wettest place 
in England" (PI. XVa, p. 202). So impressed was Miller by his first 
year's catch (of a thoroughly unexpected 152 inches) that in 1846 he 
set up additional gauges on the adjacent mountains; and for eight 
years his results were communicated to the Royal Society. While 
Miller's high level gauges suffered from over-exposure, they threw 
entirely new light on the problem of rainfall; and soon the inescapable 
conclusion was reached that the expansion and consequent cooling of 
air already near or at saturation point was the cause of the condensa- 
tion necessary for formation of rain. Mountains provide, of course, 
one of several means by which moving air can be compelled to rise, 
expand and cool. The several other processes which give rise to our 
British rainfall will be described later. 

We must not forget the great part played by little-known men such 
as Gough and Miller, Heberdcn and Hutchinson, Hoy and Barker in 
building up our knowledge of temperature and rainfall in days when 
the voluntary enthusiasm of individuals displayed the surplus energy 
arising from prosperous agriculture, nascent industry, and the splendid 
sense of freedom from external interference. Botanists and zoologists 
will recognise parallels in their own sciences; and geologists in par- 
ticular can testify to the magnificent record of those Scotsmen, Welsh- 
men and Englishmen who busied themselves with the unravelling of 
yet another aspect of the story of their own complex island. 

The other elements of weather have all received their attention. 
Luke Howard, the London apothecary, has earned his fame through 
being, in 1802, the original inventor of the present universal nomen- 
clature of clouds. He among other things strove persistently to establish 
a relation between the moon's phases and the weather, a notion which 
we now know to be erroneous. Admiral Beaufort's scale of wind force 
for use at sea ( 1 805) has received world-wide acceptance. The measure- 
ment of wind force owes much to a clergyman of Armagh, the Rev- 
erend T. Robinson, who in 1846 devised the 'cup anemometer' still 
occasionally to be found at our older observatories. And that very 
beautiful instrument, the Dines anemograph now so widely adopted 
in one form or another at meteorological stations, is also a British 



invention by one of the most eminent of those later Victorian scientists 
who so often combined mathematical ability with the capacity to 
handle and design the necessary experimental apparatus. The Dines 
anemograph registers by means of a pen on a revolving drum, a con- 
tinuous record of the fluctuations of pressure at the mouth of an open 
tube kept facing the wind. Momentary fluctuations of pressure in the 
open tube result from the variation in the speed of the wind flowing 
pist the anemometer head. These are communicated to a pen which 
in a strong wind continually rises and falls with the momentary gusts 
and lulls. A Dines anemograph in full cry during a gale is indeed an 
elegant sight and from it a record becomes available, not merely of the 
average speed of the wind for every hour of the day, but of the extreme 
speeds attained in gusts and lulls, a matter of surpassing importance to 
engineers and all who arc concerned with exposed structures (fig. 48, 
p. 152). Our knowledge of the behaviour of the lower atmosphere has 
also been vastly extended by its use. For purposes of comparison wind 
speed is in general measured at a height of 10 metres or 33 feet. 

British meteorologists too have never been forgetful of the sun; 
there are probably more sunshine recorders in these islands than in 
any other country of similar size. 'Bright sunshine' is recorded very 
simply by concentrating the sun's rays, as the sun goes round, through 
a sphere of glass on to a card ; hence the sun burns a trace on the card 
whenever it is bright enough. In general this means that the sun must 
be more than 3 above the horizon; on the most perfectly clear day the 
sun will not burn the card within about 20 minutes of its actual rising 
and setting. This type of recorder, also a British invention (Campbell- 
Stokes) has been in general use since 1880. Thermometers with 
blackened bulbs (absorbent of solar radiation) began to be used as an 
index of the intensity of 'sunshine' from i860 onward, but the proper- 
ties of one 'black-bulb' are wont to differ from the next and hence they 
cannot now be described as a satisfactory instrument for the purpose 
in view, even when enclosed, as they generally arc, within an outer 
glass tube from which the air has been exhausted. 

Much more elaborate instruments for the measurement of radiation 
have been devised in later years, some of which depend on the 'thermo- 
pile' principle; and the amount and qualities of radiation from clouds 
and from the earth as well as from the sun has been studied. At our 
major meteorological observatories (Kew and Eskdalermiir in par- 
ticular) numerous other observations arc made for special purposes 



which as far ae we can see at present arc not directly associated with 
those elements of" weather which, when averaged, go to make up 
'climate', so that they will not be mentioned here. 

The formation and behaviour of dew was first explained by Dr. C. 
Wells of London (Essay on Dav, 1818); his writings and experiments 
have been described by Sir Napier Shaw as models of scientific method. 
Following him, it was soon observed, for example by Jcnyns (already 
mentioned) in Cambridgeshire, that on a clear, calm evening die 
temperature of the dew-point is quite a fair index of the minimum 
that may be expected to occur before sunrise the next day. This remains 
a very useful rule for gardeners and fruit growers who from time 10 
time are anxious how far the temperature will fall overnight. For, as 
soon as the process of condensation begins within the layer of the 
atmosphere near the surface, sufficient heat is liberated to offset the 
further fall of temperature of the ground. On the afternoon of 4 June 
1939 at Durham the dry bulb read 81 -0°, wet bulb 58-3°; relative 
humidity was 19%; dewpoint 36 . The previous night's minimum in 
the valley nearby was 30°, and that of the succeeding night 37 . 
Relative humidities below 20% arc incidentally very rare in England; 
extreme values between 10 and 15% have once or twice been recorded. 

Among other climatic elements, the frequency of occurrence of 
snow and snow-cover, of hail and thunder, and of ground surface 
minimum temperature below freezing point are ail recorded. 'Earth 
temperatures' at various depths below the surface of the soil arc also 
taken daily and are of value to agriculturists and others, notably in 
connection with public health. 

Some mention should be made here of the instrumental recording 
of 'ground-frost', a figure subject to much misunderstanding and indeed 
abuse by journalists in search of sensational news of the weather. On 
a clear evening the earth's surface cools by radiation to outer space. 
The air adjacent to the surface follows suit; it cools pardy as a result 
of contact with the ground and partly by radiation. The temperature 
of the surface of ground exposed to a clear sky falls farther than that 
of the adjacent air; and if a thermometer bulb is placed at ground 
level it too will radiate freely and register a lower minimum than a 
thermometer in the air, assuming that the latter is itself protected from 
radiation cither to or from the bulb. Moreover the minute temperature 
differences set up owing to the varying rate of cooling of, for example, 
the different portions of a gravelly soil mean that even on the calmest 


night some slight movement and sealing of the air adjacent to the 
ground must still proceed; this again tends to keep die air a shade 
warmer than the ground. 

For purposes of comparison the minimum temperature of the air 
is generally registered by a thermometer in a Stevenson screen at the 
standard height, four feet above the ground. The effects of radiation 
from the bulb can be illustrated by putting a thermometer out on top 
of the screen; on a clear calm night such a Uiermometer will register 
a lower minimum than that in the screen which is representative 
of the surrounding air. If the thermometer is exposed on die ground 
it will commonly be found to register a minimum five, ten and occasion- 
ally even more degrees below that of the air; but a great deal depends 
on the nature of the ground and the vegetation cover, as well as the 
proximity of objects such as buildings or even fences. For purposes of 
comparison of 'ground frost', thermometers are placed with the bulb 
one inch above the ground over short grass. 

If such an exposed thermometer has registered a given minimum 
overnight it is reasonable to assume that the surfaces of the outer 
exposed leaves of plants within an inch of the ground will have fallen 
to a similar temperature. Hence it is not unusual to find that, on a 
clear night with a 'screen minimum' of 35 , the tips of the leaves of 
young potato plants just appearing above die soil may have been 
browned and that an 'exposed' Uiermometer adjacent to the ground 
may register a minimum of say 28 . 

But it is clear that the least degree of protection will check the fall 
of temperature due to radiation from objects at ground level. The 
bare branches of a tree provide such a check, as anyone can see at 
once following a night when hoar-frost has been deposited on exposed 
fields and roofs; little or no frost is visible under trees. Experiments 
have shown that even such things as thin muslin check die outward 
radiation, and many a gardener can save his young potatoes on a 
clear cold May night by draping them with sheets of newspaper. 
Moreover, as the plants grow, the upper leaves shade those below. 
Lastly, much depends on whether the soil is wet or not. A dry soil 
with plenty of air in it is a poorer conductor than a wet soil. Accord- 
ingly, heat is more readily conducted from the subsoil to die surface when 
the soil is wet, than when it is dry. Hence the biggest differences between 
minimum temperature in the air' and 'minima on ground' arc apt to 
occur when the ground is dry. and when the air also is relatively dry. 



Sensational ground-frost readings, however, bear little relation to 
those manifestations of greater significance to our daily lives, such as 
frozen pipes, ice thick enough for skating and the like. These depend 
on the fact thai, whatever the ground readings may be, the air tem- 
perature in a stratum at least some hundreds of feet thick has fallen 
well below the freezing point. Moreover ground-frost readings can 
be shown to vary quite appreciably from point to point even in a 
small garden. Hence when a London evening paper declares that 
there were "thirteen degrees of frost on the ground" at Hampstcad, 
while the minimum in the screen at Braemar was i8°, the Hampstead 
reader should not allow himself to believe that his morning nip on the 
way to the Tube was at all comparable with that associated with the 
tingling clarity of a Highland winter dawn. 

No recollection is more vivid in the writer's mind than that oi 
leaving London on a raw damp evening, then dismounting from the 
train at Blair Atholl to find a temperature of 7° at dawn on the clear 
Christmas morning of 1925. In London the sky was overcast with 
low stratus cloud and a light drift of air from the south-east prevailed. 
The air moving from the continent was rather cold, and evening 
temperatures in the town were about 35 . A depression lay in 
the Western Channel. Pressure was higher over Scotland, the night 
there was clear; there was a two-inch snow-cover in Perthshire over 
which radiation took place freely; lastly, Blair Atholl lies in a valley- 
bottom. Under such conditions temperature there fell very rapidly 
during the night and ski-ing was thoroughly enjoyable. 

It will be evident that the slightest breeze begins to stir up the 
layers near the ground and to mix them with the air above. So long 
as there is a brisk wind capable of stirring up and removing the surface 
layers as fast as they form, valleys do not become so cold as the hill- 
tops, even on clear nights. 

Some Present-day Sources of Climatological Data 

The Meteorological Office compiles and publishes summaries of the 
observations of temperature, pressure, rainfall and other forms of precipita- 
tion, wind, cloud and sunshine, from upwards of 300 stations in Great 
Britain, at almost all of which the instruments are exposed and observed 
under comparable standard conditions. Some of the results are available 
in the Daily Weather Report (including those stations which telegraph their 
records daily for plotting purposes; data are plotted on die synoptic charts 



which provide the basic material for forecasting). But the results from many 
other 'climatological' stations arc given in tiic Weekly and Monthly Weather 
Reports; these have been published since 1878 and 1884 respectively. Until 
191 1 many of the results were also summarised in the Meteorological Record, 
published by the Royal Meteorological Society from 1881 onward. We have 
accordingly quite a considerable number of British stations at which a 
stricdy comparable series of observations has been maintained for a period 
of 50 to 65 years, using Stevenson screens, modern patterns of rain-gauge 
and the like. Readers are particularly referred to the Monthly Weather 
Report (which since 191 1 has incorporated the Meteorological Record) for 
detailed information with regard to any district in which diey are interested. 

But from all that has been said in an earlier chapter it will be evident that 
comparable series of records covering longer periods are few. The strict 
reduction of the results given by old patterns of thermometer screen is a most 
troublesome proceeding, and the difficulties increase as we go farther back, so 
that even the best reductions arc not altogether free from criticism. Several 
series of observations, however, are available for consultation by those who 
wish to study the extent and character of climatic fluctuations so far as they 
are revealed by instrumental data. The dates refer to the commencing year. 

The longest officially established record, Greenwich, 1841, covers 
temperature, pressure and rainfall in great detail in the annual volumes 
oi Greenwich Observations. Rainfall data are available from 181 5 onwards. 


Oxford (Radcliffe Observatory) 1815 (Tables in Radcliffe Observations 

Durham (University Observatory) 1847 

'Edinburgh' (reduced from several 

stations) I73»-173 D ) 

1 764- 1 896) 

N.E. Scotland : Gordon Castle and other 

stations 1782- 1892. 

'Lancashire' (reduced from several 

stations) 1753 onward 
'London' (reduced from several stations) 

1 763 onward 
' Central England' 1698 onward 

Vol. 55, Appendix 1932). 
(Quart. Journ. Roy. Met. S., 

(Trans. R. S. Edin., 1897, 1900). 

(Journ. Scot. Met. S., 1893). 

(Quart. Journ. Roy. Met. S., 


(Philos. Trans. R. S. 223A, 


(Qjiart. Journ. Roy. Met. £'., 


Reduced with the aid of numerous scattered records. 
Unreduced data are available from several other stations: Orkney 1827, 
Penzance 1807 (and district), Nottingham 1810; there are also Irish records. 




'Edinburgh/, 1769-1896 

(Trans. Roy. Sue. Edin., 1900). 


Since i860 these can be extracted lrom the volumes of British Rainfall. 
Starting before i860, a table for Edinburgh (1785) is available in Trans. 
Roy. Soc. Edin., 1900; for Oxford (1815) in the RadcliJJe Observations named 
above; for York (181 1) see Q.. J. Roy. Met. S., 1933. Data for Seathwaite 
(1845) and other stations can be found in British Rain/all, 1867 and 1895, 
also Report of the British Association, 1866. Rainfall variations over England 
as a whole are reviewed by Glasspoolc in the Meteorological Magazine, 1928 
("Two Centuries of Ram, 1727-1926."). These are shown in the diagram 
on page 2O6. 

Recorded extremes ol temperature, both 'average' and 'absolute', are 
given, for a few stations, in the Book of Normals. Some of the sources cited 
above give data with regard to certain other elements such as sunshine 
duration. See also: Climatologtcal Atlas of lite British Isles. (H.M.S.O., 1952.) 

Averages of temperature and sunshine duration are now published by 
the Meteorological Office (obtainable from H.M.S.O.). The latest available 
publication covers as lar as possible the thirty-year period 1906-1935, For 
rainfall averages the later volumes of British Rainfall should be consulted. 
Other data are found summarised in many places, notably in many papers 
in the Quarterly Journal of the Royal Meteorological Society. For a number oi 
coastal stations the volumes of the Admiralty Pilot are useful, especially with 
regard to lrcquency and strength of wind; also the more recent official 
publication Weather in Home Waters (M.O. 446) which gives detailed 
inlormation for many coastal stations. Considerable caution is necessary 
before making comparisons between existing present-day stations. In 
Britain it is customary at die majority of stations to derive a mean tempera- 
ture for each mondi from the average of the daily maxima and minima, the 
instruments being read and set once daily at 9 a.m. It will be evident that if in 
a cloudy mild winter month there is a single clear cold night, the temperature 
at 9 a.m. on the following morning will be little above the minimum entered 
to the previous 24 hours. Next day at 9 a.m. the minimum entered in 
the register will be that resulting from the setting of the thermometer the 
previous morning, and thus the effect of a single cold night is to appear 
twice in die register. If die diermometers are read and set in the evening 
this difficulty is largely avoided; but it will be clear that direct comparisons 
should only be made ol the mean temperatures between stations using the 
tame hours. 



Elsewhere for the sake of condnuity old-fashioned types of screen are 
still occasionally kept in use. That at Kew has been in conunuous operation 
since 1868; but as it is attached to the wall of the building seventeen feet 
above die adjacent lawns the minima arc often several degrees higher than 
those in the standard screen. The earlier Greenwich data are also to be 
viewed with care. With regard to rainfall much has been done to standardise 
the exposure of gauges in recent years. 

The climatological observations of the frequency of snow, sleet and hail 
depend to some degree on the alertness of observers and must again be used 
with care. Remembering the inevitable difficulties attached to the assess- 
ment of eye observations a warning should be given. It is only too easy to 
make unguarded statements and draw plausible maps which will not 
stand up to analysis in respect of such material; moreover such ill-con- 
sidered compilation does not do justice to the very real efforts made by the 
Meteorological Office to standardise our records on a firm basis. 


Chapter 1. 

Brooks, C. E. P. (1929). Climate, a handbook. London, Bcnn. 

IIandisyde, C. C. (1947). The Climate of the Home. Weather, 2: 82-89 
(gives further references). 

Kendrew, W. G- (1949)- Climatology. Oxford, University Press. 
(1937). Climates of the Continents. Oxford, University Press. 

Mill, H. R. (1928). Climate of Great Britain, in Regional Essays. Cam- 
bridge, University Press, cd. A. G. Ogilvie. 

Tansley, A. G. (1939)- The British Isles and tlieir Vegetation (Chapters on 
Climate). Cambridge, University Press. 

Chapter 2. 

Brooks, C. E. P. and Glasspoole, J. (1928). British Floods and Droughts. 

London, Benn. 
Margary, I. D. (1926). The Marsham Phenological Record in Norfolk 

1736-1925. Q_.j. Roy. Met. S. 52: 27-54. 
Margary, I. D. (1927). Weather Observations at Wrendiam, Suffolk, 

1673-4. Q.,J. R°.y- Met. S. 53: 301-08. 
Merle, W. (1891). Consider aciones temper iei pro 7 annis, 1337-1344. London, 

reprod. and trans, under supervision of G. J. Symons. (The earliest 

known journal of the weather). 
I 'iioresby, R. (1826-30). Diary and Correspondence. London, 4 vols. 
Towneley, R. (1694, 1700, 1705). Account of the quantity of rain . . . 

at Towneley, Lancashire. Philos. Trans. Roy. Soc. London, 18, p. 51; 

21, p. 47; 24. p. 1877. 



. . .not now, as ere man fell 
Wholesome and cool, and mild, but with black Air 
accompanied, with damps and dreadful gloom . . . 

Milton: Paradise Lost 

It is our purpose to discuss British weather from within, and from the 
point of view of those who have to live in it, rather than to write 
a text-book of meteorology; knowledge of the physics of the atmosphere 
should be sought elsewhere, and in particular in Sir David Brunt's 
well-known works, of which the burden will be appreciated even by 
readers whose physics is rusty from disuse, or who have had little 
opportunity of study. In particular his recent book Weather Study is 
a compact and masterly introduction of the highest value to the student 
coming to the subject for the first time. It will suffice to remind the 
reader in this short chapter of a few elementary principles. 

No one can appreciate the vicissitudes of our weather unless he is 
cognisant of the characteristics of Icelandic depressions and the Azores 
anticyclone; and no one can claim full knowledge of the effects of our 
weather unless he has some acquaintance with the relationships 
between types of cloud, precipitation and the behaviour of surface air. 

The atmosphere which envelops us is in the main a mixture of 
gases ; some of these are the permanent constituents, about one-fifth 
oxygen and nearly four-fifths nitrogen with small quantities of other 
gases. Water-vapour is the principal variable constituent; carbon 
dioxide is a minor variable constituent of potential importance. Water- 
vapour is found in the earth's surface atmosphere in quantities varying 
from a minute fraction to about 4% by weight; even in the driest 
deserts it is never wholly absent. Carbon dioxide comprises, as a rule, 


only about 003%, though this figure is, of course, considerably higher 
in such places as enclosed crowded rooms. Lastly, there are minute 
amounts of solid matter as 'dust', and of liquid mostly in the form of 
extremely small water droplets; these for the most part float, or nearly 
so, in the atmosphere. Water-vapour is of overwhelming significance, 
inasmuch as its presence ultimately gives rise to the majority of the 
phenomena of weather. Most readers will be familiar with the facts; 
dry air at any given temperature can take up and retain a certain 
amount of water-vapour. Wet cloths, for example diy; puddles 
evaporate after rain. Water molecules escape from the surface of the 
liquid into the atmosphere, among whose molecules, the air being 
gaseous, they can move freely. This is the process of evaporation, 
which goes on from the surface of a liquid. 

After the process ol" evaporation into a confined space has gone on 
for some time, however, just as many water molecules will be returning 
to the liquid as are escaping from it: the air, by this time relatively 
crowded with water molecules, is now described as saturated. The 
number of molecules (or alternatively the weight of water-vapour) 
required to saturate the air rises very rapidly with temperature. 
Expressed in ordinary units, a cubic metre of air at a temperature of 
o°F. will, when saturated with water-vapour, hold almost 1 gram. 
But at 32°F. this figure rises to 487 grams; at 6o°F., 12-91; at 8o°F., 
nearly 25 grams; and 100 °F., 50 grams. Frequently, instead of express- 
ing the amount of water-vapour in grams, the vapour pressure is 
stated ; that is, that part of the whole pressure of the atmosphere which 
arises from the water-vapour within it. 

In general, however, air out-of-doors is not saturated; the atmos- 
phere contains less water-vapour than the maximum permitted at 
that temperature, so that evaporation from damp objects or from a 
water surface can still take place. The ratio between what the air 
does contain and what it can contain ai the same temperature is called 
the Relative Humidity and is customarily given as a percentage. 
Taken over the whole year lor example, the average relative humidity 
near sea-level in England is about 80%. At any given temperature, 
the lower the relative humidity, the more rapidly evaporation can 
take place. 

Evidently if air containing water-vapour is cooled, a temperature 
will in lime be reached at which the air is saturated; if the cooling is 
continued, some of the vapour will no longer be retained as such; it 



will condense into the liquid form, or direcdy into the solid form if 
the temperature at which condensation is initiated is below freezing- 
point. The surplus as liquid will either appear as very small drops in 
the atmosphere, or will be deposited on neighbouring objects as dew 
(liquid) or hoar-frost (solid); though it may be added that if the 
temperature of the ground on which dew has been deposited sub- 
sequently falls below the freezing point, the frozen dew-drops are also 
described as hoar-frost. If the very small drops suspended in the 
atmosphere are numerous, they will impede visibility and present the 
appearance of cloud or fog. The droplets in cloud (or fog) are com- 
monly of the order of 0-02 to 005 mm. or ^ to ^ inch in diameter; 
the rate of fall of such small droplets is almost imperceptible and hence 
they are carried along in the stream of air. They may increase in size 
by further condensation upon diem, or may diminish by evaporation. 
Either process however, is often rather slow; hence cloud formed over 
say, Devonshire is quite capable of being carried to Oxford and 
beyond given the right conditions. 

The so-called dust, in the form of particles so small diat many 
thousands are normally present in a cubic centimetre, is largely swept 
up from parts of the earth's surface such as deserts and areas of loose 
dry soil; a fact known to our Fenland farmers as well as those of 
Oklahoma. Some are composed of the particles of salt left after the 
evaporation of sea-spray; some may be of volcanic origin; some arisr 
as smoke from imperfect combustion. The diffusion of such small 
particles through the surface layers of the atmosphere greatly affects 
visibility, or the degree to which objects can be clearly seen and 
recognised. The variations due to the drift of smoke from large in- 
dustrial towns are only too conspicuous in countries such as Britain. 
Visibility at times is seriously limited at a distance of 60 miles down- 
wind from London, and instances have been known of visibility being 
affected as much as 200 miles from the source of the atmospheric 
pollution. In lands lying nearer desert areas, when in the afternoon 
there is plenty of surface turbulence, extensive dust-haze develops in a 
layer which may be as much as 8-10,000 feet deep, a fact well known 
to those R.A.F. men who were engaged in Libya. 

Condensation as Fog or Mist 
Having recognised the effects of water-vapour and also of dust, we 


may consider the various methods by which air can be cooled on a 
sufliciendy large scale to provide widespread condensation. 



1 F 

< > 


065O OT03 


TOP OF F06 / 

• top OF FOS 



1 F 



Fio. 3 
Temperature Gradient at intervals during the dissipation of a September 
morning fog. Based on observations by Hcywood, Q,. J.Roy. Met. S., 1931 

With light stirring of the air after sunrise, the lowest temperature in each 
instance is above the ground but below ihc lop surface of the fog 

In the first place, this may arise from contact; when warm moist 
air rests on or flows over a cooler surface condensation of moisture at 
and near the surface commonly occurs, either as dew on the ground, 
or mist or fog in the air. For example, when damp ground is losing 
heat on a calm evening by radiation to a clear sky, some evaporation 
of moisture takes place at first into the thin surface layer of air 
immediately adjacent. But at the same time this layer is no longer 
being stirred up and removed, but is being furdier cooled in contact 
"ith the ground, and so the moisture from it later condenses on die 



cold ground as dew. There is no doubt thai the formation of dew 
is considerably facilitated if the ground is already slighdy damp. 
After a spell of extremely dry weather it is often found that very little 
dew is deposited. The dewy mornings of a normal English September 
owe much to the fact that the ground is normally rather damp at that 
season as a result of August rains. The approximate correspondence 
between the dew-point and the minimum air temperature recorded 
on a clear calm night has already been mendoncd (p. 18). 

Moreover, if the wind continues to blow with some vigour during the 
evening, the surface layer of cool and almost saturated air as it forms will 
be stirred at intervals and dispersed into the slightly warmer layers above. 
Hence a windy night is not in general marked by much formation of dew. 

In the intermediate case, however — a very slight stirring of the 
surface air — the cooling of the surface layers soon leads to condensation 
on the ground; but also if the overlying air is already close to satur- 
ation, and more moisture is carried up from the surface by the slightly 
turbulent air, the result is that condensation is initiated in the air as 
well as on the ground and fog results. The mixing process between 
adjacent layers of nearly saturated air plays its part and as soon as 
fog droplets are formed they themselves act as radiators, cooling the 
adjacent air. The depth which such a fog attains depends on many 
factors, some of which arc mentioned later in this chapter. It can in 
any case be shown that almost anywhere in Britain the undulations of 
the ground, the varying rate at which different surfaces cool, and the 
varying degree to which the ground is sheltered or not, all lead to 
slight local movements of air sufficient on many occasions to initiate 
the process of fog formation and accumuladon, especially in the valleys 
and pockets of our varied countryside. Indeed such movements are 
necessary; air is a poor conductor, and were it to remain absolutely 
still it is considered that a night fog could not exceed about four feet 
in depth. 

Bearing in mind that autumn months are often rather rainy, thai 
with the waning power of the sun the temperature and the rate of 
evaporation are decreasing by comparison with summer, and that 
the nights are lengthening, we can readily see why morning mists 
and fogs are characteristic of quiet mornings towards the end of 
die year. 

Further, on a clear night the air over long coarse grass tends to 
fall to a lower temperature than over short grass, or among trees. 


The blades of grass are themselves radiators, and the dps are relatively 
well insulated against conduction from the roots. Hence the effective 
cooling surface is greater where the grass is coarse; this in turn cools 
the air among the grass blades. Hence we observe mists developing 
earlier, and lasting longer over damp marshy uncultivated hollows. 
Lack of drainage not only affects the moisture content of the adjacent 
air; it is often (bund that if the water cannot drain away, the movement 
of the dense cold air is also hindered. In a later chapter we shall see 
that the consequent ponding of cold air and development of frost- 
hollows and frost-pockets has very important effects. 

The term 'fog' is used (by international agreement) whenever the 
visibility is less than one kilometre or 1,100 yards. With this criterion 
there is no doubt that Tacitus was right; by comparison with many 
other countries Britain is foggy, though not more so than Holland or 
Belgium or N.W. Germany. In this country we have adopted the 
term 'thick-fog' for visibility less than 220 yards, i.e. sufficient to 
provide an impediment to road and rail traffic and to impress the 
average London, Glasgow or Manchester citizen. Fog formed in the 
manner described above during a night of clear skies, is appropriately 
known as 'radiation fog'. In recent years careful studies have been 
made of the factors governing its development and the varying depth 
which may be attained overnight. It is very evident diat the rate of 
variation of the moisture content of the air up to two thousand feet 
or thereabouts is important; in particular conditions favouring fog 
development occur when the moisture content of the atmosphere per 
unit volume slightly increases with height in the first two or three 
hundred feet. It is quite possible for this to happen in a moist air 
stream without its being quite saturated. But as soon as the layer at 
the ground is cooled below saturation point, the air a few feet higher 
is so near saturation that the very slight mixing due to turbulence is 
enough to cause the mixture to be saturated. Hence the fog, which 
begins to form as the familiar thin creeping layer over open damp 
spaces, quickly grows in thickness; and within an hour traffic is seri- 
ously impeded. In the East Midlands it is not uncommon for a night's 
radiation fog in November to attain a depth of five hundred feet. 
Moreover, a layer of fog is itself a radiator, and a partial rellector of 
any warmth it receives from above by day. Hence when the sun is 
low it is often unable to dispel a thick overnight fog, and if a second 
calm clear night follows the fog may well deepen further. Fortunately 



it is rarely that two or three days in succession remain sufficiently 
calm for this to occur. For the London fog, Dec. 1952, see p. 56. 

It will now be evident diat with a slighdy more rapid decrease 
with height of the moisture content of the atmosphere in the lower 
layers, there will be many occasions when at any given level the 
mixing of the lower layer with that above will not quite result in satur- 
ation of the mixture. Fog will then not form, or will only begin to 
form after the temperature has fallen for some hours, say towards 
dawn; the balance is delicate, and careful measurements of the local 
humidity of several levels are recorded for forecasting. The importance 
of being able to forecast the times at which surface fog would begin 
to develop, and the depth it would attain, was very great during 
the war on account of the needs of returning R.A.F. pilots. Much 
of this recent investigation is due to Mr. W. C. Swinbank, one of 
the younger physicists who began his studies at the University of 

In the light of what has been said regarding the approximate 
correspondence between the overnight minimum and the dew-point 
on a clear calm night it may be asked, how does the temperature 
continue to fall under conditions of dense fog; that it can do so, 
although rather slowly, is well known. The answer appears to reside 
in the fact, already mentioned, that outward radiadon continues 
from the top of the layer of fog, and the cooled air from the top of the 
fog sinks to the ground level. The fall of temperature is however, 
comparatively slow. As the substance being cooled is mobile, the 
cooler elements at the upper surface as they begin to sink are con- 
tinually being replaced by warmer elements from below. 

If a dense fog occupies a valley with snowy uplands on either 
flank arising above the top of the fog, the air cooled by radiadon over 
the uplands continues to flow down into the valley and extremely cold 
days result, as the fog is likely to prevent any penetration by the 

If the valley itself is also deeply snow-covered fog is not so likely 
to persist. Petterssen has shown that this arises from the fact that the 
saturation vapour pressure over ice is less than that over water. 
Expressed otherwise, if ice is evaporating into a dry atmosphere, 
saturation will be readied more quickly than if supercooled water 
were evaporating into air at the same temperature. Otherwise, if 
air is cooling towards saturation and there is a snow cover, the water 

4^ ** 


. '.,-** 

Plate II: Alto-cumulus illuminated by the setting sun. 
London: July 1936. 

//. Rail Kerr 


vapour in the air will begin to condense on the snow-surface and so 
the moisture content of the air will remain just too low for conden- 
sation as liquid drops. Hence at low temperatures the presence of a 
snow cover often tends to prevent the formation of radiation fog in 
the air above. We thus see a reason why the Scottish Highland 
valleys are often deeply covered by snow, and very marked 'inversions' 
exist, that is, pools of exceptionally cold air occupying the valley- 
bottom; yet there is little or no fog. 

But if the damp air 
overlying the snow is at 
a temperature above the 
freezing point, i.e. the 
snow is wet, the air is 
cooled by the snow and 
saturation may well be 
reached at a tempera- 
ture above the freezing 
point. In diis case con- 
densation begins as 
liquid drops in the air, 
and a fog quickly forms 
which gradually falls to 
a temperature close to 
32 °, that of the thawing 
snow beneath. 

Indeed, fogs over 
thawing snow in Britain 
are common, but al- 
though they can develop 
in the manner above, 

for the most part they are due to the advection of air; in this case 
the movement of a moist air current from a warmer source. 

For we must recognise that moist ah in motion, passing over a 
cooler surface, may also be sufficiently cooled to produce fog: and that 
this can, and does occur even if the sky is overcast. The best-known 
example in the world of persistent and frequent 'advection fog', as it 
is called from its mode or origin, is that of the Newfoundland Banks. 
Just to the east and soutii-cast of Newfoundland the sea surface 
temperatures are those of the Labrador current, and are very low for 

C.B.S. D 

Fig. 4 

Advection fos with smoke effects in Midlands 

and north east; 22 October 1937 i.sec p. 5) 



the latitude. Yet the abnormally warm waters of the Gulf-Stream 
Drift lie only a short distance farther south. Hence almost any move- 
ment of air from points between south-west, south and cast brings 
moist warm surface air on to a much cooler surface, and fog results. 

It is to be noted that movement of air implies some degree of 
turbulence over the surface; packets of saturated air in which some 
condensed droplets are already to be found are therefore carried 
upward from, and downward to, the surface. Fog, therefore, may 
quickly attain a considerable depth; much depends on the strength 
of the wind, the relative humidity of the air above the chilly surface 
layer and the extent and suddenness of the temperature difference 
between the warm and cold waters. Sea-fogs of this type may range 
from only a few feet in depth, so that the masts of ships are intcrvisible, 
to many hundreds of feet. We owe to Sir Geoffrey Taylor of Cambridge 
one of the most elegant discussions of the factors governing the depth 
and density of such fogs, in the Quarterly Journal oj the Royal Meteor- 
ological Society ibr 191 7. 

Advection fog and mist occur quite frequently in the British Isles. 
As we have seen the needful conditions are provided whenever the 
ground surface is cold enough, and especially when the country is or 
has recently been covered with snow and a mild damp south-westerly 
wind spreads inland. Moreover, moist air approaching our south and 
east coasts in spring from the direction of the Continent must frequently 
cross rather cooler water as it approaches our shores. The effects on 
our weather are very characteristic and will later be discussed. 

Condensation of some of the water-vapour is likely to occur 
in a belt along the boundary between two nearly saturated currents, 
as we have already seen in the last chapter. On a small scale this 
goes far to explain the development of radiation fog, rather than mere 
dew. On a large scale, some might argue that a belt some miles wide 
of 'horizontal mixing fog' would appear to be formed when for 
example, a slow-moving moist but cool current from the continent, 
and a warmer but also very moist current from the ocean converge. 
But it must be remembered that the amount of moisture condensed 
due to this process is very small and unless it is confined to a 
very narrow layer near the ground it does not suffice to make a fog, 
largely because pure horizontal mixing is unlikely. The warmer 
current tends to override the cooler; we get low cloud rather than 
surface fog. 


With a light southerly wind in the English Midlands such a state 
of affairs is not uncommon in autumn and early winter. Moreover, 
smoke from the towns drifts along in the colder surface layer and cuts 
down the visibility. If the converging air streams are more lively, 
forming a minor warm front, drizzle or rain from the cloud above falls 
into the chilly moist air below and brings it to saturation point, so 
that a belt of 'frontal fog' is then found. Over the damp, retentive 
clays of die Midlands it is quite rare to find good visibility in autumn 
and winter unless there is a strong wind. Advection fog due to a 
warm southerly airstream creeping inland over the cooler ground; 
valley radiation fogs deep enough to be farther spread by a light wind 
over the neighbouring country; and the occasional frontal fog are all 
reinforced by smoke. Hence, as a whole, the area most subject to mist 
and fog lies in a long belt from London to South Lancashire and 
Yorkshire, with patches of greatest frequency over our great inland 
cities, and local upland districts such as the High Chilterns which may 
stand above some of the radiation fogs. Allowing for such local 
exceptions, throughout this belt fog, that is visibility less than 1,100 
yards, is observed on upwards of 50 days yearly. (Cf. Durst, p. 55). 

Some local condensation also takes place at the base of the atmos- 
phere if a very cold current of air blows over a much warmer water 
surface. This produces the same effect that we observe in the 'steaming' 
of the surface of a hot bath; locally on the surface the cool air is first 
saturated and condensation begins. But the saturated air has not to 
move very far before it is again mixed with the unsaturated air above ; 
hence the 'steam' continuously forms along the surface, but as quickly 
dissipates. We may often observe the steaming of unfrozen lakes, 
rivers, and ponds in frosty weather. Apart from such slight effects, 
however, 'steam-fog' has no significance in Britain ; but on the Canadian 
Great Lakes an outburst of Arctic air early in winter may sometimes 
give trouble through this cause to the local shipping, at the season 
when the waters are still ice-free. 

The differing behaviour of our atmosphere with regard to fog 
formation plays its part in scenic effects. Under English conditions 
the combination of a deep snow cover with a temperature well below 
freezing point and clear skies is rare. In the Scottish Highlands it is 
more common and Dr. Fraser Darling reminds us in his Natural 
History in the Highlands in this series how men and animals alike enjoy 
the brilliant frosty days following a heavy snowfall. Under such 



conditions a rarely-seen amethyst colouration is perceptible in the 
landscape and especially in the shadows on the snow, given the really- 
pure atmosphere of those fortunate areas far removed from towns. 
It is probably attributable to the fact that the air is so pure and free 
1'rom dust particles that even the blue wave lengths are not so much 
scattered as usual. (Cf. chapter 8, p. 146.) 

Normally when the land is snow-covered and the sun shines from 
a cloudless sky, shadows are blue in colour because they receive all 
their illumination from the brilliant blue sky, largely reflected by the 
snow. Exceptionally brilliant cold days early in March 1947 will long 
remain in the memory of many country dwellers, and Plate 29, p. 206 
will remind us of the unforgettable splendour of the Derbyshire moors 
on that occasion. But in England at least it is much more usual to 
find that our occasional bright winter day with the more normal 
snow-cover is at the same time decidedly hazy (Plate 11, p. 94). This 
is largely on account of the fact that there is generally a surface 
inversion some hundreds of feet deep in which smoke-haze is penned, 
seriously limiting the visibility in combination with the light mist 
that so often drifts from neighbouring districts in which less snow has 
fallen, or where it has already melted. Hence the brilliance of mid- 
February in the upper Gudbrandsdal, let alone Pontresina, is very 
rarely matched in England, though it sometimes develops for brief 
spells in the clearer air of the Scottish Highlands. Can we ascribe to 
such influences the marked regard of the Norseman for brighter colours? 
Although his descendants in our north-western dales show in the style 
of their barns — as in many other respects — the traditions of their fore- 
fathers, those who observe the fell farms of Lunds from the northbound 
Scottish express as it crosses Aisgill beyond Garsdalc will see no sign 
of that cheerful Scandinavian red. From Reynihals the valley of Litli 
Langadalur (Wrynose and Little Langdale in the Lake District) is 
indeed beautiful ; fell and dale alike bear witness to climatic action of 
the kind that has moulded Norway; but barn and wall remind us of 
the fundamental difference in our winter climate, which nowadays but 
rarely gives us a sample of the dazzling snowy brightness of the North. 

Condensation as Ci.oud 

The visible results of condensation of water-vapour in the form 
of cloud are evident to us all. The droplets which go to form cloud 


arc: of the same order of magnitude as those in fog, i.e. they 'float' 
rather than fall. With sufficiently low temperatures the liquid droplets 
arc replaced by minute ice crystals; in general, however, as we shall 
sec, clouds in British latitudes arc not wholly composed of ice crystals 
until we attain levels above 20,000 feet. 

Air, as a mixture of gases, obeys the laws of gases according to 
which the pressure, volume and temperature of a gas arc closely 
related. For example, if we take an enclosed volume of air and allow 
it to leak through a small hole into an adjacent enclosure from which 
the air has been partially extracted, we shall find that the pressure of 
die air on the side of the first vessel is lowered, but also that its tem- 
perature will be lower. Air consists of large numbers of molecules in 
constant motion; with expansion of the air into the adjacent vessel, 
the decrease of molecular activity is manifest to us as a fall of tem- 

A mass of air rising from the earth's surface expands owing to 
the decreased pressure upon it of the air above; and provided that it 
is not saturated with moisture, it falls off in temperature by 5-4° F. 
for each 1,000 feet that it rises. Normally air contains a good deal ol 
water- vapour, though not enough to saturate it; but if it continues 
to rise it will sooner or later cool to a temperature at which it is 
saturated, and condensation will begin; the small drops forming what 
we call cloud. We shall see that there aie occasions when moist air 
rising from the surface is checked in its ascent; if the check takes place 
before the rising air has cooled to saturation point, cloud will not form. 

Masses of air can be caused to rise from the earth's surface in three 
principal ways. First, by heating of the surface layers to such a tem- 
perature that the rate of fall with height or 'lapse-rate of temperature' 
exceeds the figure given above. Secondly, by the movement of the 
air over obstacles such as hills or mountain ranges, or movement ol 
lighter air over denser air in its path, or by convergence of air currents 
differing in density. Thirdly, by turbulence; vigorous movement of 
air over the earth's surface is accompanied by considerable friction, 
and on a windy day packets of air from the surface are continually 
being carried upward and downward in a layer many hundreds of 
feet thick, if not more, as a result of the disturbed (low. 

Clouds form with little difficulty over the British Isles for several 
reasons. Fundamentally of course the surface air has only too fre- 
quently moved within the previous few hours from over die sea and 


is generally moist, that is, it has a high relative humidity. Little 
cooling is therefore needed to initiate condensation. The cooling of 
the air may be brought about, as we have already seen, by contact 
with a cooler surface beneath, in which event we get fog. Or it may 
be brought about by ascent and consequent expansion; and it follows 
that the more humid the air the lower the cloud base is likely to be 
when the air ascends. 

Ascent may be the result of the heating of the surface layers 
adjacent to the ground; this is die normal process on a summer 
morning, forming cumulus clouds. If the process is sufficiendy developed 
so that the ascending moist currents boil up to a level at which 
further condensation takes the form of ice crystals, the top of the cloud 
acquires a "fibrous" appearance (PI. 20a, p.127 ) and is called 'cumulo- 
nimbus'; this is the cloud in which thunder-sTiowcrs so often occur. 

We may add a little more on the heating of the surface layers. This 
may be effected by the heating of the ground, which heats the air in 
contact with it; but also if cold air flows over a warmer surface, similar 
effects are produced. These are very important over our seas in winter. 

Within an area of low pressure there is necessarily some convergence 
of air-stream towards the centre and hence ascent; so that areas of low 
pressure frequently tend to be cloudy quite apart from the additional 
factors leading to production of cloud discussed in Chapter 4, p. 63. 

Just as fog once formed itself loses further heat by radiation, 
so a cloud sheet towards nightfall may thicken as a result of cooling 
due to radiation from its upper surface. Thin sheets or layers of cloud 
('stratus' types) can sometimes be attributed to this cause, especially 
when they appear rather quickly in a clear sky. We should distinguish 
the types of cloud associated with the rapidly rising currents due to 
surface heating as 'cumulus' and 'cumulo-nimbus'. 

Dissipation of Cloud 

It follows that descent of air leads to compression and warming; 
and the descent or subsidence of saturated air with the consequential 
rise in temperature means that the air will no longer be saturated. If 
cloud was present, it disappears. Evidence of the slight descent of air 
after crossing quite small ranges of hills is frequently given by the 
existence of a gap in the clouds to leeward. For example, when low 
stratus cloud arrives on a south-west wind on the Welsh coast, it is not 


uncommon to find extensive breaks in the cloud sheet over Shropshire. 
We shall sec that much depends on the height of the hills and the 
height of the clouds. Hills do not, in general, affect the flow of the air 
to more than about three times their own height. We shall see, too, 
that some descent of air is a necessary consequence of die existence ol 
a region of higher atmospheric pressure, or an anticyclone, so thai 
there is an a priori tendency for widespread clear skies; in a later 
chapter wc shall discuss the extent to which in Britain this remark 
needs amplification. 


It is extremely important to bear in mind certain additional 
properties of small water droplets once they are formed. In the first 
place, provided they remain very small, they can be cooled far below 
the freezing-point without turning into ice; but when drops are thus 
super-cooled, a very small disturbance such as the impact of the wing 
of an aircraft or even the movement of a cyclist or walker causes them 
to freeze immediately. The finest cloud droplets which are normally 
present can be cooled to about zero Fahrenheit before freezing. We 
therefore find that the great majority of clouds seen in Britain up to an 
altitude of 20,000 feet are still principally composed of water droplets; 
but above this level we can assume that with rare exceptions, high 
clouds are composed of ice crystals in the form of minute hexagonal 
crystals or plates. 

Confirmation of the composition of the thinner cloud-sheets is 
often forthcoming. If the sun or moon shines through a thin veil of 
'cirro-stratus', i.e. a sheet of icc-crysLal cloud generally above 20,000 
feet, some of the rays are refracted by their passage through the crystals. 
To an observer on the earth these rays appear to arrive from a position 
in the sky 22 in angular distance from sun or moon. Hence a narrow 
ring of light is seen to surround the luminary, often forming a complete 
circle when the veil of cloud covers a large area between the sun or 
moon and the observer. Not uncommonly the ring is seen with the 
colours of the spectrum more or less developed, the ice crystals behaving 
as small prisms with regard to those rays passing through them. 
Occasionally a variety of additional phenomena appear including 
mock suns (or moons) and haloes of larger radius; for a full discussion 
the reader should refer to writers on atmospheric optics. 


When shining through a thin layer of cloud at a lower level, 
however, the sun or moon generally appears surrounded by faindy 
coloured rings (coronae) close to the luminary; these arise from 
diffraction or the scattering of light by the 'grating' of small water 
droplets Uirough which the rays are passing. It will be observed that 
these appear even through a thin layer of alto-stratus at heights of the 
older of 10,000 feet in winter, confirming diat although the temper- 
ature at that height is very far below die freezing point very small 
liquid droplets still predominate in the composition of die cloud. 

But if water-vapour condenses at a temperature below 32 G F. it 
is liable to take the form of ice crystals. We can see, therefore, that if 
we have an ascending mass of saturated air iu which condensadon 
began at say 45°, as the air rises and cools degree by degree its vapour 
content for saturation must decrease. Condensadon will continue; and 
the liquid drops so formed may be carried up by the rising air to levels 
at which the temperature is far below freezing. But that part of the 
condensation which takes place below 32° will be directly as ice; and 
so we find that in the upper part of a large rapidly ascending current, 
such as is manifest in a big cumulo-nimbus cloud, ice crystals and 
water drops are found together; finally, at very high levels, ice-crystals 
predominate. This can often be observed; cumulus cloud retains the 
rounded edge characteristic of water-drop cloud, until it grows beyond 
a certain level, after which the top assumes a cbaracterisdc iibrous app- 
earance. The ice crystals are swept out by the wind, just as they are in 
the well-known high cirrus 'mares' tails' we so often see; and this fibrous 
structure at the top of die high-piled towers of cloud is the distinguish- 
ing mark of 'cumulo-nimbus'. The predominance of ice crystals in 
the upper part of the cloud arises on account of the fact that the 
vapour pressure required for saturation over ice is less than that over 
water. Hence in an atmosphere containing ice crystals and super- 
cooled water drops, further condensation takes place on the ice crystals 
which dius tend to grow in numbers and size. If the growth is rapid 
they will begin to fall; and it has been argued that the greater part of 
the precipitation as rain falling from clouds in temperate latitudes 
arises in litis manner, becoming rain at lower levels. For if a cloud is 
merely composed of minute water drops, it is by no means clear why 
they should begin to aggregate. Nevertheless, it is known that rain- 
drops can furm in clouds in which no ice whatever can possibly exist; 
hence the problem of the actual formation of raindrops widtin a cloud, 


simple though it may appear, is not yet fully explained. It is known 
that other processes may have to be invoked at temperatures above 
50T. In a turbulent cloud the droplets formed as a result of con- 
densation at different levels are carried upward and in the same pan 
of the cloud droplets differing considerably in size may be found. U 
this is granted, then momentarily the vapour pressure over the bigger 
drops will be less than that over the smaller drops; hence further 
condensation is likely to take place on the bigger drops, which con- 
sequendy increase in size. Moreover clouds in which there is consider- 
able turbulence appear to be essential if rain is to fall. Hence it was 
also thought that coalescence of the droplets by collision was an 
adequate explanation; but this will not do, because the small droplets 
themselves 'float' in the turbulent air and there are many very tur- 
bulent cumulus clouds from which rain does not fall. Thus we need 
further explanation. Quite recently some physicists have expressed 
their belief that even at lower temperatures there are frequently not 
enough ice crystals present to initiate extensive growth of raindrops, 
and yet it rains. There is much room for further study of the properties 
of water in die atmosphere at low temperatures, as the results already 
obtained by the research group under Professor G. M. B. Dobson at 
Oxford have already shown. Reference should also be made to recent 
papers in the Quarterly Journal by F. H. Ludlam and B. J. Mason. 



High Clouds. 
Mean Lower 
level 20,000 ft. 

Cirrus (Ci) 

Cirrocumulus (CC.) 

Cirrostratus (Cs.) 


Detached clouds of delicate and 
iibrous appearance, without sha- 
ding, generally white in colour, 
often of a silky appearance. 

A Cirriform layer or patch com- 
posed of small white flakes or of 
very small globular masses, with- 
out shadows, which are arranged 
in groups or lines, or more often 
in ripples resembling diosc of 
die sand on die seashore. 

A diin whitish veil, which does 
not blur the oudincs of die sun 
or moon, but gives rise to halos. 




Middle Clouds. 
Mean Upper 
level 20,000 ft. 
Mean Lower 
level 6,500 ft 

Low Clouds 
Mean Upper 
level 6,500 ft. 
Mean Lower 
level close to 
the ground. 


Form Description 

Altocumulus ^Ac.) 

A layer (or patches) composed 0/ 
laminae or rather flattened glo- 
bular masses, the smallest ele- 
ments of the regularly arranged 
layer being fairly small and thin, 
with or without shading. 

Striated or fibrous veil, more or 
less grey or bluish in colour. 

A layer (or patches) composed o) 
globular masses or rolls: the 
smallest of the regularly arranged 
elements are fairly large; they are 
soft and grey, with darker parts. 

A uniform layer of cloud, resem- 
bling fog, but not resting on low 

A low, amorphous and rainy 
layer, of a dark grey colour and 
nearly uniform. 

Thick clouds with vertical de- 
velopment; the upper surface is 
dome-shaped and exhibits 
rounded protuberances, while 
the base is nearly horizontal. 

Heavy masses of cloud, with 
great vertical development, 
whose cumuliform summits rise 
in the form of mountains or 
towers, the upper parts having 
a fibrous texture and often 
spreading out in the shape of an 

Classification of Clouds 

Luke Howard, the London apothecary, writing in 1802 was first 
to name the characteristic types of cloud ; he distinguished cumulus 

Clouds with 
Mean Upper 
level 20,000 ft. 
Mean Lower 
level 1,600 ft. 

Altostratus (As.) 
Stratocumulus (Sc.) 

Stratus (St.) 
Nimbostratus (Ns.) 

Cumulus (Cn.) 
Cumulonimbus (Cb.) 


(heap cloud), stratus (or layer cloud), cirrus (the high wisps or streaks) 
and nimbus (from which rain falls). These terms have gradually been 
combined and enlarged and the present-day international classification 
of the main types of cloud is universally adopted. For the reader's 
convenience they are tabulated here with some details of their occur- 
rence and characteristics; a fuller discussion will be found if desired in 
most meteorological textbooks. The highest clouds in the British Isles 
are very rarely above 40,000 feet. The normal limit, formed by the 
base of the stratosphere, is about 33,000 feet. 


Moisture may be precipitated from the air in the form of drizzle, 
rain, snow, sleet or hail, which fall; or dew, hoar-frost and rime which 
arc directly deposited on exposed surfaces. 

Drizzle is nowadays distinguished from rain inasmuch as drizzle 
droplets, being small, drift down on the wind rather than fall. In 
general, a drizzle drop is of the order of 5," to +' in diameter, 
i.e. of the order ten times the diameter of a cloud-drop. Raindrops 
range from about £" to {" diameter. Any drop of water of larger 
size, if it forms, is quickly torn apart again due to its own limited 
surface tension and to frictional effects as it falls through the air. 
Further, the resistance of the air imposes a maximum velocity on falling 
raindrops depending on their size; for a drop of \' diameter this is 
about 25 feet per second, or 18 m.p.h. It follows that no rain can fall 
if the 'vertically' rising currents exceed this figure. Not uncommonly 
with vigorous convection on a humid summer morning when vertical 
currents of the order of 30-40 feet per second prevail, cumulus 
and cumulo-nimbus continue to grow for some time until at length the 
first large drops begin to fall. They immediately cool the ground and 
the lower air by evaporation ; the vertical currents are then further 
diminished and the rain is suddenly released in a way known to all 
Englishmen who watch cricket-matches. 

Snow occurs when the minute ice crystals resulting from con- 
densation below the freezing-point fall; at very low temperature (below 
zero Fahrenheit) single ice crystals fall in the form of a very fine dust. 
At higher temperatures complex aggregations of crystals develop in the 
infinite variety of patterns we know as snowflakes and when the tem- 
perature is close to freezing-point, those may in turn stick together so 



thai the snowflakes become quite large. Snowflakes will only reach the 
ground if the temperature of the air through which they fall remains 
cold enough. Upon the ground there may sometimes be appreciable 
accumulation of wet snow with the air temperature at 34% but it soon 
melts unless the temperature falls. The persistence of a snow-cover is 
greatly favoured if the ground surface is already below freezing point. 

Snow which has partly melted in its fall, so that some of die drops 
fall as rain, is known as sleet. As English winter temperatures fre- 
quently lie in the neighbourhood of 40 ° at sea level, precipitation there 
is most commonly observed as rain. But a climb of 1,000 feet on a 
rainy and windy day with saturated surface air means that temper- 
ature will fall to about 36-5°, at which some flakes of melting snow 
arc likely to be observed mixed with rain. A further climb of 1,000 
feet is likely to bring the air temperature down to 33 , at which heavy 
wet snow is most probably falling ; and if the temperature falls below 
32% dry snow flakes will be swept before the wind and drifting will 
occur. Clearly one of the most characteristic features of the British 
winter climate is the rapid increase with altitude in die frequency with 
which snow or sleet is observed either to fall or to lie. 

There are, however, occasions when although the surface air 
remains below freezing-point, raindrops fall from a warmer layer 
above. Such raindrops are slightly supercooled and freeze immediately 
on impact, and if the surfaces of objects at ground level are already 
chilled below the freezing-point the accumulation of 'glaze' as it is 
called may occasionally be serious. In January 1940 enormous 
damage was done to trees, telegraph wires and transmission lines by a 
phenomenal ice-storm of this kind throughout the West .Midlands; rain 
fell for several hours with an air temperature in the region of 28 (fig. 5). 
Similar events occurred in Sussex in March 1947 (Plate 13, p. 98). 

It appears that a small raindrop carried through a stratum of air 
below freezing-point may freeze to form a small hail pellet and around 
it supercooled cloud droplets freeze on impact. As some air is included 
the resultant aggregation appears opaque. It is probable that in cool, 
rather stormy winter weather raindrops at moderate altitudes are 
often carried just sufficiently upward in die turbulent air to freeze, 
and the small opaque pellets of 'soft hail' result. Soft hail is common 
enough on our coasts in winter, especially farther nordi, and can often 
be attributed to the slight uplift and disturbance caused when a cool 
stream of air, usually from a westerly point, passes over the land. 



iooe =j% 

Such air streams tend 
to be more frequent 
and to blow more 
strongly in more north- 
ern latitudes, as we 
might expect. Hence 
at those observatories 
at which an alert 
watch is kept, soft hail 
is observed to fall on 
about 10 days yearly in 
the south, 15 days 
yearly in Lancashire, 
upwards of 20 in West- 
ern Scodand and at 
Aberdeen and over 30 
in Lewis and the 
Shetlands. No precise 
comparison of the inci- 
dence of soft hail can 
however be made as 
its recording depends 
so largely on the alert- 
ness of observation. 

True hail, as it is often called, is much more frequent in summer 
and falls from thunderclouds. The violent vertical currents developed 
in a big cumulo-nimbus cloud carry the raindrops formed near the 
base of the cloud to great heights when they freeze ; at the same level 
ice crystals are found and some of these accumulate on die outside of 
the pellet of ice. The pellet may then fall and partly melt; but only to 
be swept up again and re-frozen. Sometimes the repeated up-and-down 
tossing results in the formation of hailstones of considerable size; 
when these are cut, the 'concentric' structure often is revealed. In 
England hailstones die size of marbles are regarded as large; in this 
country the limit for single stones so far reported is about the size of 
a golf ball. Bearing in mind that the region in which such violent 
upward and downward currents prevail in a cloud is often quite small, 
we can at once see why the occurrence of really large hailstones is 
somewhat capricious and fortunately limited to very small areas. The 

Fie. 5 
Glazed frost, 28 January 1940; note blizzard at 
Dalwhinnie. Gla/c, from rain falling at tem- 
peratures below freezing point, from Wiltshire 
to the North Wales border; very heavy snow- 
fall followed in Lancashire and the north-west 



extent of convectional currents depends a good deal on the extent to 
which surface heating is effected by comparison with the temperature 
of the air at higher levels; hence hailstones often attain greater violence 
in countries nearer the Equator, especially when they comprise 
plateaux at a high level such as the South African veldt. The West 
Suffolk storms of July 1946 did so much damage to crops that an appeal 
fund was launched. True hail is liable to do considerable damage to 
standing crops, and sometimes to greenhouses and the like. On the 
whole it appears that severe damage by hail tends to be associated 
with those districts in which summer thunder is most frequent. Inland 
Kent and Sussex, East Yorkshire, Nordiamptonshirc, Suffolk and 
Worcester have all figured in reports of serious damage by hail in 
recent years. In the past one of the worst storms on record occurred 
at Cambridge in August 1843. Damage in the town was then estimated 
at £25,000; by to-day's standards this might perhaps be multiplied by 
ten. The hailstones lay almost knee-deep. 

Lightning and its accompaniment thunder, which is a sound effect 
caused by the setting up of air waves by the passage or a spark, are also 
associated with violent convection in cumulo-nimbus cloud. As a 
result of electro-static effects arising from the brcaking-up of larger 
raindrops, perhaps also from their passage through the air, and from 
friction among the ice crystals, different parts of the cloud become 
highly charged and spark-discharges occur between them, or from 
cloud to earth. The sheet lightning low down towards the horizon, 
so often seen on a warm and cloudy summer night, is mainly attribu- 
table to the reflection of distant flashes from the under surface of the 


Little has been said with regard to the processes by which the small 
cloud-drops become the much larger drizzle— or rain-drops— and begins 
to fall. It was at one time considered that aggregation by collision in 
the turbulent rising currents might be a sufficient explanation, but this 
view is not wholly acceptable, as we saw on p. 39. We may, however, 
observe that many of the drizzle- or rain-drops in a cloud arc consider- 
ably larger than those which on the average go to form the cloud 
itself; and, however the process of aggregation is effected, the formation 
of drizzle and rain only occurs when there is considerable turbulence 
within the cloud. Coagulation of droplets by collision may be possible 
if some of the falling drops are already of larger size. In particular, a 
great deal of the formless layer cloud with which we arc provided 


gives no precipitation. It can be shown that widiin such cloud, tur- 
bulence in general is slight, and in particular vertical ascent of the 
condensed moisture is prevented above a certain level so that the cloud 
is generally limited in thickness; and that such a cloud is probably 
homogeneous, i.e. all the droplets are of the same size. 

With regard to the formation of the deposits of dew, hoar-frost and 
rime, dew and hoar-frost have already been discussed (p. 18). 

Rime is frozen fog; we have already seen that fog below freezing- 
point is in general still composed of liquid drops. Cycling through 
such a fog, and even walking, impacts the supercooled drops so thai 
they at once turn to ice on one's coat, while motorists quickly note the 
deposit on the windscreen of a car. The slightest wind deposits the 
fog as minute frozen ice crystals against the sides of fences and the 
like, and gives us a small-scale reminder of the processes which lead 
to icing on aircraft. Rime-deposits associated widi persistent low cloud 
and strong wind on our mountain summits are often very conspicuous. 
On Ben-Nevis 'frost-feathers' sometimes grew outward to a length of 
five feel on the exposed masts carrying the instruments in the days of the 
old Observatory. The scenic effects associated with widespread rime- 
deposit on trees are often extremely beautiful. 

Before we end these reminders of the part played by our vaporous 
atmosphere in producing 'weather', we may also recall that water- 
vapour has important effects with regard to the transmission of radia- 
tion. Large quantities of vapour in the atmosphere act to some extent 
as a screen especially with regard to the long-wave terrestrial radiadon, 
that is outward radiadon at night; hence the greatest daily ranges of 
temperature occur when the air for many thousands of feet above the 
surface is decidedly dry (p. 172). 

Stability and I.NSTABrLiTY of Surface Air 

Having reviewed the pari played by water-vapour in regard to the 
formation of cloud and precipitadon we must now discuss its sig- 
nificance with regard to 'stability' as we call it, in the atmosphere. 
The term is used with reference to die equilibrium of a small mass of 
air. Normally in air which is not saturated die rate of fall of pressure 
from the earth's surface upward, consequent on the decreased weight 
of the air above, is such that a mass of air rising from the earth's 
surface, expanding and cooling as it does so, would fall in temperature 


by 5 4°F. for each 1,000 feet of ascent. This is known as the 'dry 
adiabatic lapse rate', the term adiabadc implying that no further heal 
is transferred to, or from the rising mass of air on its way; while 
'lapse-rate' is a convenient term for the lapse or decrease of temperature 
with height. On a clear and breezy afternoon in summer the rate of 
fall of temperature in the disturbed air near the surface approximates 
very closely to the dry adiabatic for two or three thousand feet above 
the ground. Typical daily variations are shown in Fig. 6. 

Let us suppose that on a given occasion the rate of fall of temper- 
ature with height for several thousand feet above the ground is 4°F. 

per 1 ,000 feet. Such a prevailing lapse 
rate is shown in simple form in Fig. 7, 
p. 48, plotting temperature against 
height. Consider now what happens if 
a small mass of air adjacent to ground 
level is warmed by five degrees. It will 
then be less dense than the surrounding 
air at the same level, and will accord- 
ingly rise. When it rises its temperature 
must fall off by 5-4° for each 1,000 feet 
of ascent, assuming for the moment 
that there is no further heating of the 
rising mass due to some external cause. 
By the time it has risen 3,000 feet 
(approximately) its temperature will 
have fallen to the same value as that of 
the surrounding air at the same level, 
and as it will then have the same 
density as the surrounding air it will 
not tend to rise any farther. Due to its 
momentum it may rise a little farther, 
but its rate of ascent will quickly 
diminish and come to a stop. 
On the other hand if the prevailing lap*.e-rate in the air as a whole 
were say, 6° per 1,000 lect, a rising "bubble" from the surface (and 
5 warmer, as above) would find itself only 5-4° cooler after 1,000 
feet whereas the environment would be 6 a cooler. That is, the air 
which was warmed 10 begin with now finds itself on attaining a level 
of 1,000 ft. relatively warmer, lighter, and more buoyant than the air 


Fio. 6 
Diurnal variation of Stability 
over land. A, early morning; 
B, midday; c, evening; broken 
line, dry adiabatic (based on 
Pelterssen, by courtesy o( 
Messrs. McGraw-Hill) 

Plate III: a. Anticyclonic weather; inversion fog in a Somerset valley, 


b. From the same view-point, free from fog. 
W. G. V. Balchin and N. Pye (from the Qiiarterly Journal of the 
Royal Meteorological Society). 

^ n 

The Times 
R.A.F. Crown Copyright Resentd 



surrounding it at the same level, and so, not only will it continue to 
rise but it will tend to do so more quickly than at the surface. 

Suppose now that instead of heating calm surface air as above, 
the surface air is in vigorous motion; a strong wind is blowing. As the 
air moves rapidly over the irregular surface of the ground, slight 
obstacles are continually causing small masses to rise. If in the air 
as a whole the lapse-rate is greater than 5-4° per 1,000 feet it is evident 
that a small mass of air deflected upward by an obstacle will con- 
tinue to rise (as in B above). But if the lapse-rate in the air as a whole 
is less than the critical value of 5-4° per 1,000 feet a small mass of air 
deflected upward will find itself cooler than its environment and will tend 
to fall back to the surface from which it rose. In the first case the air is 
in unstable equilibrium ; a small displacement is followed by further 
displacement. In the second case the air is in stable equilibrium ; a small 
displacement is at once followed by a return to the original position. 

It will thus be evident that the meteorologist must take into 
account the prevailing lapse-rate throughout the lower atmosphere 
before he can decide whether rising air currents due to surface heating 
will come to a stop near some particular level as (A) above, or whether 
they will continue to rise. In the event of turbulent flow over the sur- 
face he must decide whether a sufficient degree of heating or cooling 
will also take place as the air moves to cause the air to become more 
stable or more unstable (cf. die paper by R. M. Poulter cited at the 
end of this chapter). 

Further, the stability or odierwise of an air current is complicated 
by the existence of water-vapour. Rising air cook and may reach a 
level at which the water-vapour begins to condense. 

Just as the process of evaporation, that is the conversion of a 
substance from the liquid into the gaseous state at the same temper- 
ature, requires heat, so the process of condensation liberates heat. 
Suppose therefore that the temperature of the air at the surface is 

Plate IVa: Summer afternoon clouds over the English Channel; an inlra-red 

photograph from 20,000 feet. The line of cumulus clouds formed a Utile way inland 

where the sea breeze rises is a frequent development when the general direction of 

the wind lies along the coasts. Distant alto-stratus with some cirro-stratus above. 

b: View over N.E. France looking westward from 26,000 feet, June 1944. 

Lower cloud sheet breaking into normal summer cumulus below. Distant frontal 

cloud spreading over southern England. 

C.R.S. E 



50 and the moisture content is such that the relative humidity is 
80%. Ii air at a temperature of 50 is saturated it contains 9-33 
grammes per cubic metre of water-vapour; hence the sample in 
question must contain 80% of this, i.e. 7-46 grammes. We find from 
the tables that this is the amount appropriate to saturated air at 
43-7 F. Therefore if a packet of this air rises from the surface its 
temperature will begin to fall at the expected dry adiabatic lapse-rate 

6000 1 





30 40 SO 

Temperature \'F) 

30 40 SO 

Temperature (*P ) 

Kio. 7 
Diagrams illustrating the behaviour of rising air masses up to 5,000 feet 
UJt. Warmer unsaturated air at C is 5 warmer than its environment A; 
it can rise to B. right. Rising warm air from C reaches saturation at D; 
as it is still warmer than its environment it will continue to rise, and its tem- 
perature will fall as lor saturated air along D-E 

of 5-4°F. per 1,000 feet, and a temperature ol 43 ^'F. will be attained 
at about 1,200 feet; at this level, therefore, the air will be saturated. 
Should the air continue to rise, expand, and cool, some of its moisture 
must continually condense. But as the process of condensation liberates 
heat, this will be communicated to the rising air mass, and the resultant 
rate of fall of temperatures will be considerably less than 5-4° per 
1,000 feet; at the temperatures given above the lapse-rate for saturated 
air is about 2-8° per 1,000 feet rise. On account of the fact that very 
warm air holds much more moisture, when condensation begins the 
amount of heat liberated is greater; hence at temperatures round 90° 
the saturated adiabatic lapse-rate has decreased to about 2 per 1,000 
leet, while at zero it is nearly 4 per 1 ,000 feet. , 



It will therefore be evident from die adjacent diagram (fig. 7) that 
if a warm bubble of air rises in an atmosphere in which the lapse-rate 
is represented by AB, and die warm bubble has initially a temperature 
C, much depends on whether saturation is reached. If the rising air is 
initially moist and saturation is reached at a low level (D) the tem- 
perature of the rising packet of air will decrease along a line DE so that 
it will remain wanner than its surroundings and continue to rise. This 

He - oht(Ft) 

6000 -1 

4000 ■ 

Height (Ft) 




SO 40 -<-J 

Temperature V F) 
Inversion at B 

Saturation reschcJ 
here locally 


40 50 

Temperature ( Fj 
Inversion at B 
Fto. 8 

Diagrams illustrating typical behaviour ol rising air masses up to 5,000 feet 

•eft. Cloud, if formed, does not ascend above F. right. Cloud, if formed, 

ascends rapidly to great heights. Saturated air at F is still considerably 

warmer than its environment given by B-G 

is the normal result of convection over heated ground when the air 
supply is moist. Indeed on occasions the rise becomes so rapid that 
die vertical currents assume the dimensions of a gale. Glider pilots 
and others, caught in the centre of large cumulus clouds, have at times 
reported rising currents of 50-60 feet per second even in Britain. 
Abroad, where humid air masses at high temperatures are most 
common, the very gradual fall of* temperatures represented by the 
saturated lapse-rate means that on many occasions the difference of 
temperature between the rising air mass and its surroundings tends to 
increase with height and the uprush of air becomes even more violent. 
This fact helps to explain die greater height reached by cumulo- 
nimbus clouds in regions such as the Southern United States and the 



greater intensity of all the associated phenomena — lightning, hail, 
violent up-currents and the like. 

In practice our notions of die behaviour of rising air 'bubbles' 
should be modified to some extent to allow for some mixture along the 
boundary as they rise. Further, once convection is established on a 
line day the rising bubbles, breaking away from the surface every few 
minutes, arc replaced by descending air which is in turn heated from 
below. Hence a circulation is set up which explains why the develop- 
ment of cumulus cloud does not take place uniformly over the whole 
sky, but in patches which arc often rather evenly distributed; between 
them, the areas of blue sky represent the areas in which descending air 
currents prevail. (Cf. PI. 23, p. 138). 

But the rising bubbles of humid air, which go to form cumulus 
cloud above the line at which saturation is reached, do not invariably 
rise to a great height. Suppose that the air at a higher level, is subsiding 
— and it is the meteorologists' business to detect any such tendencies 
with the aid of his daily upper-air sounding balloons — such subsiding 
air becomes warmer due to compression and the result may well be 
the establishment of a temperature inversion. Diagramatically we 
should then get a result as Fig. 8: the temperature of the environment 
being represented by ABC with an inversion at B, the lapse of tem- 
perature in the rising bubble height be shown by DEF, with cumulus 
cloud having its base at E. In this event the rising bubble will find 
itself at the same temperature as the environment at the point F. If 
it were to climb farther, it would immediately be colder dian its 
surroundings and accordingly must sink back. 

We very frequently find that the upward growdi of cumulus clouds 
is checked in this way, especially when Britain lies in the westerly air 
stream on the northern flank of a summer anticyclone. In such cir- 
cumstances the surface air stream is fairly moist and is warmed up in 
the daytime over the land. But as subsiding air is characteristic of 
anticyclones we frequently find that a slight inversion, sufficient to 
check the upward growth of cumulus, is present at a height of die 

Plate 5 

a. Lake District: broken strato-cumnlus in lurbulenl air over mountains at 
sunset. October. Westerly wind, maritime polar air. 

». Hertfordshire: tailed and plumed cirrus, September evening. Wisps probably 
indicative of strong wind at high levels with eddies forjning the "plumes" 

Gjrti \iu-brrrr 

B. .1. Croat* 


order of 3,000-4,000 feet. Hence the trequent sight of the fleets of 
■fair-weather cumulus' — so-called because we can be sure that if the 
cumulus clouds show no great vertical development there will be no 
chance of showers. On the other hand, if once they reach the ice- 
crystal level — shown by the tell-tale fibrous appearance of the uppci 
edge — as towering cumulo-nimbus, showers may at once be expected. 
The part played by the ice crystals is mentioned earlier (p. 38). 

F10. 9 

Flow of air over hills 

left. Stable atmosphere, cloud rising 10 no great height over hills an,! 

generally smooth outline of upper surface, right. Unstable atmosphere, 

cumulus cloud rising 10 a much greater height. Compare Fig. 8, p. 49 

It will be evident that if the warm air at a higher level is replaced 
by colder air, the likelihood of an unstable atmosphere in which 
vertical currents, once initiated, can ascend to great heights unchecked 
is greatly increased. The diagrams above are simplified, and the actual 
state of the atmosphere as shown by upper air soundings is often much 
more complicated. It will be evident that there will be many occasions 

Plate 6 

u. Essex: late rammer sunset, September. Daytime strato-cumulus decreasing and 
settling down to patchy stratus. 

b. Axlkstkre, Derbyshire: summer sunrise, August. Underside of alto-cumulus 
•beet illuminated by the rising sun, still below the horizon; red rays predominate 
through surface haze. Higher and more distant cirro-stratus already fully lit in 

dear sunshine. 

Cyril Xrwbtrry 



when the atmosphere will remain stable if the rising air at any given 
level remains unsaturated; but if, locally, saturation is just reached 
the resultant instability may be marked. 

Moreover, with a moist wind blowing the effect of a mountain 
range is frequently to cause much more cumulus development over the 
summits than over the plains below. Nothing is more characterisdc, 
with a light westerly wind on the northern margin of a High, than the 
clusters of cumulus among and above the hills of Snowdonia or the 
Lake District while over the sea and the coastal plains die sky remains 
clear (Plate 32, p. 223, of the Lcvcn-Kent estuary). 

"Over the smooth sands 
Of Lcven's ample estuary lay 
My journey, and beneath a genial sun 
With distant prospect among gleams of sky 
And clouds, and intermingling mountain tops . . ." 
(Wordsworth, The Prelude). 

This can be shown to be partly due to the lifting of the whole block 
of air, and partly to the localised heating of the ground surface cither 
in constricted valleys, or at higher levels on the uplands. Instability is 
indeed more likely to develop over healed upland, for in sunny weather 
the ground headng is just as great as it would be in the plains, while at 
the higher level the prevailing atmospheric environment is cooler. For 
this reason some of the Western Pennine valleys in which unstable 
humid air from the south-west finds itself heated and constricted show 
rather a high frequency of summer thunderstorms (cf. p. 264; also the 
paper by Sir Charles Normand cited at the end of diis chapter). 

It is not intended diat this chapter should swell into a meteorological 
text, but a brief mendon should be made of another characterisdc 
climadc feature so frequently superimposed on any Bridsh prospect, 
namely, the tendency to grouping and alignment of the fair-weather 
cumulus cloud which is so common. Cumulus growth in calm weadicr 
can be regarded as die result of a series of 'cells' in each of which 
ascending air in the centre is complemented by descending currents 
on the exterior. If a breeze is blowing the clouds drift gendy over the 
country. But if at the level of the top of the clouds the wind direction 
is appreciably different in speed and direction from diat at the surface, 
the clouds dispose themselves in different patterns; sometimes these 
take the form of 'cloud-streets' or longitudinal rolls aligned roughly 


in the direction of the surface wind, but at others they lie transverse 
to the wind. On other occasions a sheet of cloud breaks up as a result 
of convection set up within it into a series of rounded masses indicating 
that a vertical circulation has been set up in the cloud sheet, leading 
to regions of ascent (where the masses of cloud are thick) and descent 
where the cloud is thin, or absent. Assemblages of evenly-disposed 
cloudlets whedier at low or at high levels, in the form of more or less 
globular masses are very familiar to all who observe our British skies. 
Other familiar cloud formations arise, as we shall sec, in association 
with 'waves' forming along the boundary between upper and lower 
streams of air. For a detailed account of the recent theories of forma- 
tion of some of our familiar cloud-patterns reference should be made 
to papers by Sir Gilbert Walker and Sir David Brunt, in the Qiiarterly 
Journal of the Royal Meteorological Society, 1 932 and 1 937. 

The development of longitudinal cloud streets under convection 
is of especial importance to the glider pilot. Providing as they do 
continuous lift over a long stretch of country, with this aid nights of 
many miles can successfully be made. 

At this point the reader may well ask whedier in extreme instances 
any limit is ultimately imposed on die vertical ascent of moist air from 
the surface and the consequent upward growtii of cloud. There is 
indeed an upward limit imposed at die base of the stratosphere. 
Disregarding small inversions, which in general are not more dian a 
few hundred feet thick, it is universally found that temperature 
decreases with height up to a level which averages ten miles above the 
surface at die equator, nearly seven in Britain (33,000 feet) and about 
four miles over the Poles. Above that height balloon ascents, and now 
aeroplane ascents, reveal that the temperature remains stationary or 
begins slowly to rise. The region in which the temperature as a whole 
falls with height is called the troposphere; that above is the strato- 

Rising air masses cannot climb through an inversion (p. 49) and 
as far as the troposphere is concerned the base of die stratosphere 
fulfils the same function. The maximum height to which moist air 
from the surface can rise is dius determined ; and for practical purposes 
no cloud whatever is observed in the stratosphere. Indeed recent 
investigations by Professor Dobson and his team at Oxford go to show 
that practically no water- vapour is present in the stratosphere at all. 
The height of the base of die stratosphere is known to vary; it is 


generally higher over anticyclones and may attain 43,000 feet or so. 
This may be regarded as the greatest height at which any cloud is 
ever to be seen over Britain. The tops of the highest cumulo-nimbus 
may occasionally tower to nearly 30,000 feet. Hence, on a clear 
winter day with unstable polar air, the tops of high-piled cumulo- 
nimbus arc characteristically visible around the horizon at great 
distances, upwards of a hundred miles being common. They arc thus 
a common ingredient of die stormy winter sunsets over the sea. 

For it will now be evident that convcctional activity is not confined 
to the land on warm summer days. In winter it is the sea that is the 
source of warmth, and instability is liable to develop as we shall sec 
whenever a colder air stream blows across it. Convcctional activity 
100 is not confined to the daytime under such conditions, hence 
towering shower clouds are equally to be seen at night when in winter 
the hard north-wester from Greenland roars over the Hebrides behind 
a vigorous depression. 

Those whose interest has been aroused and who wish to pursue 
these problems of the physics of the atmosphere may be commended 
to a variety of reading matter. In addition to Sir David Brunt's 
delightfully clear introduction to meteorology {Weather Study, London, 

[son 1 941), in which more advanced texts are named, readers of 
different tastes may appreciate the following recently published 
works, all of which are generally obtainable. 

(i) Kjjndrew, W. G. (1949). Climatology. Oxford, University Press. A 
clear and comprehensive text book of wide appeal. 

(ii) Miller, A. A. (1950). Climatology. London, Methuen. A compre- 
hensive introductory text especially addressed to the university 
student; the latest edition contains valuable additional matter. 
BoTLBY, Cicely M. (1948). The Air and its Mysteries. London, Bell, 
latest edition. A charmingly concise and most informative smaller 
book which will be found very acceptable for the armchair. 

(iv) Bilham, E. G. (1938). The Climate of the British Isles. London, Mac- 
millan. The well-known standard rdsum6 of the material col- 
lected under the auspices of the Meteorological Office, with 
informative tables and diagrams; an indispensable reference. To 
this add die Climatological Atlas oj t/ie British Isles (M.O. 488: 
H.M.S.O. 1952). 

(v) Brooks, C. E. P. (1949) Climate through tlte Ages, (second ed.). 
London, Benn. The standard work of its kind in English; an 
invaluable digest of an enormously wide subject. 


(vi) Brooks, C. E. P. (1950). Climate in Everyday Life. London, Benn. 

(vii) Brunt, D. (1939). Physical and Dynamical Meteorology, Cambridge, 
University Press. For more advanced students widi a sound 
mathematical and physical background. Cited repeatedly in 
die meteorological literature of every country in the world. 

(viii) Kimble, G. T. (1951). The Weallier London, Pelican Books. An 
exemplary introduction. 

(ix) Hare, F. K. (1953). The Restless Atmosphere. London, Hutchinson. 
Recommended; compact and stimulating, especially for students. 

To these may be added two well-known comprehensive Service 
text-books published by the Stationery Office: — Admiralty Weather 
Manual (A. G. Forsdyke); Meteorology for Aviators (R. C. SutclifFe), 
together with the invaluable Meteorological Glossary and The Weather Map 
obtainable from the same source; these last are still astonishingly good 
value at five shillings each. Coast-dwellers will appreciate Meteorology 
for Seamen (C. R. Burgess, 1950. Brown and Ferguson, Glasgow). 


Brewer, A. M. (1946). Condensation Trails. Weatlicr, 1: 34-40. 
Brunt, Sir David (1937). Natural and Artificial Clouds. Q.J.Roy. Met. 

S. 63: 277-88. 
Darling, !•'. Frascr (1947). Natural History in the Highlands and Islands. 

London, Collins' New Naturalist. 
Dobson, G. M. B. (1946). Temperature of the Upper Atmosphere. 

Weather, 1: 58-65, 73-77, 115-22. 

(1949). Ice in the Atmosphere. Q,. J. Roy. Met. S. 75: 117-3°- 
Durst, C. S. (1940). Winter Fog and Mist Investigation in die British 

Isles, 1936-7. Meteorological Office (M.O.M. 302). 

(1949). Meteorology of Airfields (M.O. 507). H.M.S.O. 
Heywood, G. S. P. (1931). Wind structure near the ground and its 

relation to temperature gradient. Q_.J. Roy. Mel. S. 51: 433-55. 
Howard, Luke (1803). On the Modifications of Clouds. London. 
Ludlam, F. H. (1951)- The production of showers by the coalescence of 

cloud droplets. Q_. J. Roy. Met. S. 77: 402-417. 
Normand, Sir C. (1938). On instability from water vapour. Q.. 1. Roy. 

Met. S. 64: 47-68. 
Petterssen, S. (1941). Introduction to Meteorology. New York and London, 



Poulter, R. M. (1938). Cloud Forecasting: the Daily Use of the Tephi- 
gram. Q.-J- Roy. Met. S. 64: 277-92. 
(1946). The Depth of Forecasting. Weather, 1: 137-40. 

Schumann, T. E. W. (1938). The Theory of Hailstone Formation. 
Q.J. Roy. Met. S. 64: 5-20. 

Siieppard, P. A. (1947). The constitution of clouds and formation of 
rain. Science Progress, 35: 185-92. 

Simpson, G. C. (1941). On the Formation of Cloud and Rain. Q..J- Roy- 
Met. S.67: 99-133. 

Smith, K. M. (1946). Forecaster's Progress: the Teplugram. Weather 1: 

Swinbank, W. C. (1943). Synoptic Division Technical Memorandum 52. 

Meteorological Office, London. 
Taylor, G. I. (191 7). The Formation of Fog and Mist. Q,. J. Roy. 

Met. S. 43: 241-68. 
Walker, Sir Gilbert (1933). Clouds and cells. &. J. Roy- Met. S. 59: 


Note on the London Fog, December, 5-9, 1952 

This four-day log, during which the temperature in the Thames 
Valley remained persistently near freezing point, gave rise to peculiarly 
unpleasant consequences. Much throat irritation, bronchitis and 
pneumonia developed, with a sharp rise in the death-rate, especially 
of the elderly. Discussions fbcussed on the probable accumulation of 
sulphur compounds, among others, in addition to the coarser pro- 
ducts of combustion. (Cf. Q. J. Roy. Met. S. 80. 261-278). 





Be patient, swains; these cruel-seeming winds 

Blow not in vain. 

Thomson : The Seasons 

Our islands occupy a northerly latitudinal position in which there 
is a great difference between the power of the summer sunshine 
and that of winter, arising largely from the much lower angle of 
elevation of the sun's rays above the horizon in winter and partly from 
the lengUi of die day. At that season the rays must traverse the atmos- 
phere by a much longer path with the result that a greater proportion 
is absorbed or scattered on die way. To this we may add that much 
of the remainder is reflected from the upper surface of the extensive 
low cloud of winter. Even with clear skies we must also at times allow 
for reflection by a snow surface of much of the incident radiation during 
the day, and consequently littie gain of temperature by the adjacent 
air compared with the nocturnal loss. Over the earth as a whole 
about half the total solar radiation incident upon the atmosphere 
reaches the earth's surface. Over open countiy that part which has 
penetrated the atmosphere is very largely absorbed in a diin layer 
adjacent to the ground surface. Because of this absorption of the 
warmth in a thin layer the air in contact with the ground is also 
warmed; whereas at night the loss of heat by outward radiation from 
the surface is accompanied by cooling of the adjacent air. Some of 
the outgoing and incoming radiation is absorbed by the air; on 
balance under normal conditions the temperature of the surface air 
reaches a minimum just before sunrise, and a maximum about two 


hours after noon. This lag is due to the finite time necessary for the 
heat to be transferred upward from ground to screen level. On the 
Eiffel Tower in summer the maximum is normally recorded aboui 
4 p.m. 

Associated with the daily rise ol temperature and the consequent 
stirring-up of the surface air the average wind speed at ground level 
also tends to attain a maximum early in the afternoon. This daily 
stirring-up goes far to explain why, in higher latitudes where there is 
little winter sunshine, the daily range of temperature at ground level 
is still noticeable. The diagram below (fig. 10, p. 60), showing the 
normal diurnal range of temperature at Pare St. Maur in Paris and at 
the top of the Eiffel Tower (984 ft.) shows how great a part is played 
by surface heating and cooling. Similar effects are found on isolated 
mountain summits, for example on Ben Nevis. 

With us the observed daily range between maximum and minimum 
depends not only on the march of the sun; it is also affected by changes 
in the source of the air, and by the cloudiness or otherwise of the sky 
and the extent to which the surface layers continue to be stirred up at 
night. In winter the cumulative effects of such changes .ir<: sometimes 
large enough to make the maximum for the 24 hours occur during 
the night. The range of temperature is also much lower by the sea, 
where the air is much more likely to be in motion as a result of the 
temperature difference between sea and land. The day-to-night 
variation of sea surface temperature is almost negligible compared wiih 
that of the land, largely because water is semi-transparent to much of 
the solar radiation. Hence the heating effect must be spread over a 
considerable depth rather than a mere few inches. Statistics show that 
at small islands in die Hebrides where cloud, wind and maritime 
location combine their effects the average range between the daily 
maxima and minima is about half that at inland stations. 

In Britain, as the tables show, the average daily range varies 
between about 6° in December and 12 in June at island and coastal 
stations, and about io° in December to 20° in June at inland stations. 

Plate 7 

a. Tanf.ra Bed, Summer Isles. N.W. coast ol" Rail lllifC" May. Fine clear sky 

above, patchy slrato and stratus-cumulus developed over the sea. 

b. F.XMOOR, Somerset: March. Cumulus nearly becoming cumulo-nimbus. 

Unsettled shower)- weather in unstable manlinv polar air. 


jnttiti b'alur 

II. .1. Ciaalh 


Cyril .V 




On ihc great majority of days in these islands the air to a depth 
of many thousands of feet is moving more or less rapidly. To us, 
therefore, it is in general more important to know where the air has 
come from; and afterwards to consider what the sun, for example, will 
be capable of doing to it when it arrives. The temperatures we 
experience on any given day of summer or winter depend very largely 
on the origin of the air and its life-history, that is, on what has happened 
in that air on its way to our shores. So far as our perceptions are 
concerned the effects of solar heating or nocturnal cooling are generally 
secondary in importance to those arising from the qualities of the air 
masses we receive. 

We must, therefore, consider why these islands are relatively 
breezy, and why they arc subject to such marked and frequent changes 
in the qualities of the air. This involves a summary of the average 
distribution of surface barometric pressure. On our rotating globe with 
its surface composed of land and sea and subject to isolation through 
a partly absorbent atmosphere, our atlases remind us that an average 
disposition of surface pressure arises such that pressure is generally 
high in the central Atlantic and low from South Greenland to North 

It will at once be evident that we in the British Isles he in winter 
on the south-eastern side of the region in which low pressure pre- 
dominates near Iceland. To the eastward, pressure rises steadily in 
the direction of the 'great Siberian High'; this is one of the most 
prominent features of any pressure map of the Northern Hemisphere 
between November and March. In the central Atlantic between the 
Azores and Bermuda the average pressure is also higher than it is 
to the north and south. 

In summer this 'Azores High' occupies a more northerly latitude 
and becomes more emphasized by comparison with the average 

Plate 8 

a. Dunstabij! Downs, Bedfordshire: April. Gliding site of thr- London Gliding 
Club. Fine April afternoon with detached cumulus bounded by an inversion 
above. Moderate N E, surface wind blowing from the cool North Sea. Tem- 
perature 59" 

b. Ail psTBBH, Derbyshire: AugUSI afternoon. Cirro-stratus above stralo-cumulus, 
heralding the approach of a depression. Note how the decreased convection, 
due to the oncoming cirro-stratus, has led to the degrading and flattening of 

afternoon cumulus. 



pressure over the continents in the same latitudes; and while the 
'Icelandic Low' is still present, it is much less marked than in January. 
To the eastward, pressure is low over south-west Asia throughout the 
summer. ' 

*£.T.-. Jaly 

X ShVauri Jan. 

4 8 10 12 
Time iu Hours 


Fio. io 

Diurnal ranijc of temperature at Pare St. Maur Paris, and at the 

top ol the Eiffel Tower (from Lake's Physical Geography, by courtesy 

of the Cambridge University Press) 

Owing to the earth's rotation air in motion in the northern hemi- 
sphere is subject to a force, the so-called 'deflecting force', propor- 
tionate to its velocity and acting at right angles to the right of its path. 
If then the force on the air arising from a difference of pressure, and 
acting from high to low pressure is exactly balanced by the deflecting 
force, these two forces must act in opposite directions so that the air 
will move along the isobars with low pressure to the left. The greater 
the pressure force, i.e. the closer the isobars, the greater the deflecting 
force required for balance, and hence the greater the wind speed. It 
is in fact found that the winds of the free atmosphere at about 2,000 
feet around Britain conform quite closely to this balance. 



At the surface however, the movement of the air is less rapid owing 
to friction and the deflecting force and pressure force are not so well 
balanced, hence while we observe that the direction of the surface 
wind lies somewhat across the run of isobars, the deviation as a rule is 
of the order of io° over the sea, and 30 over the land. 

At 2,000 feet over the sea or an open low-lying plain, however, 
the frictional effects are less, and both the speed and direction of 
movement of the air closely approach those which would prevail in 
theory. Hence if we observe the surface wind direction in the open 
country on a breezy day, we shall find that the motion of low clouds 
at 2-3,000 feet level is steadier and more rapid than that of the surface 
wind, and from a direction a few degrees farther round the compass. 

The effects of friction can be illustrated by the rough rule that in 
windy weather, the surface wind at sea blows at about two-thirds the 
speed of that at 2,000 feet; inland among trees and buildings the 
average speed may be less than one-third of this. Occasional gusts, 
however, approach the velocity recorded in the free air above. With 
fairly closely spaced isobars and a west wind of Beaufort Force 8 at 
2,000 feet (40 m.p.h. i.e. gale force) we may find a mean wind speed 
over an hour of about 25 m.p.h. (between force 5 and 6) at, say, 
Blackpool; and 15 m.p.h. (force 4.) at typical inland stations. But 
occasional gusts even inland may attain 40 m.p.h. and it will readily 
be understood that the fluctuations in wind speed, between gusts and 
lulls, are greater and more frequent at the inland station. 

The considerable variations in the mean wind speed at British 
stations arising from exposure produce marked results with regard to 
vegetation, and hence greatly affect the appearance of the landscape. 
From the point of view of our own perceptions these variations are 
extremely noticeable, and go far to explain the varying climatic 
reputation of many places of resort. Examples abound; the traveller 
on a northbound express on the old North-Western route cannot but 
notice the gradual change in the attitude of exposed trees, particularly 
as the line quits the relative shelter of South Cheshire in the lee of 
Wales for the windier Lancashire plain north of the Mersey. If from 
Leeds he takes the old Midland route through the Pcnnines, the roar 
of the south-wester as the train crosses the exposed viaduct of Ribble- 
head goes far to explain the struggling ash beside the fellside farm. 
We can be sure that in the eighteenth century the beauty of Wethcral 
Woods flanking the deeply incised Eden was the more appreciated by 



north-bound travellers alter the miles ol battering rain on the top of 
a coach over Shap. And to-day no Cambridge man will deny the 
significance of wind across the Fcnland by contrast with the shelter 
of a great city, when on a February night the searching north-caster 
scours die station platlorm as he dismounts from the London train. 
Neither will Edinburgh men forget the bleakness of Carstairs Junction 
when on such a night they too are required to leave the Glasgow- 
bound express. 

Variations in the strength ol the surlace wind arise from many 
other features. The trend of water inlets such as the Solent (fig. 44, p. 147), 
the Bristol Channel, or the Firdi ol Clyde leads to some 'canalising' of 
the flow of air which must be allowed for by all whose business takes 
them into the neighbourhood of the water. Steep slopes and summits 
experience additional strength in the wind irom particular directions. 
There is lor example, .1 peculiar bleakness even in the northern 
Chiltems near Whipsnadc, on days when a strong north wind crossing 
the snow-covered Midlands besets those exposed slopes. Much may 
depend on die detailed local contour ol the hills; most Ice slopes arc 
sheltered, but some are surprisingly harried by wind. A noteworthy 
example is the south-west slope of the Crossfcll escarpment ; given the 
right conditions the north-caster blows down this treeless scarp witii 
exceptional local strengdi, as the widely known 'helm wind' (see p. 148). 

All these effects arise as a result ol die varying amount of distortion 
and frictional disturbance of the great streams of air crossing the 
country; more must be said about the origin and characteristics of 
these air-streams. 

The average distribution of pressure outlined above indicates at 
once diat for a great part of the year air will flow over these islands 
more or less in accordance with the average isobars, namely Irom 
south-south-west in mid-winter, west-south-west in summer. More- 
over, the average pressure gradient, with pressure falling from S.E. 
to N.W., is much greater in winter; hence we might correctly deduce 
that stronger winds tend to prevail then. Bui the frequency with 
which the wind comes from all other points ol the compass besides 
those mentioned, leads us to recognise at once the extent to which 
the systems of pressure indicated above are only the 'average' of a 
very large number of daily readings of pressure taken over many 
vears. From day to day the distribution of pressure and the direction 
and strength of wind can vary very widely from the 'average state ol 



affairs' shown above, so that while far more often than not our air 
supply has recently crossed a wide stretch of sea it does not necessarily 
arrive from south-westerly points. 

Barograph records soon reveal that pressure during a whole month 
fluctuates considerably; and instead of merely considering average 
pressure for the month, we should instead give more attention to the 
results of the movement of those regions of high and low barometer 
which give rise to the barometric fluctuations we observe and which 
we call anticyclones and depressions. Hence we can first embark on a 
short review ol the characteristics of the majority of depressions in our 

part of the world. 

If, lor a moment we consider the average distribution of pressure 
shown in the diagrams as the result of planetary conditions it is clear that 
in a region like the North Atlantic very different types of air will 
frequently be juxtaposed. Air which has followed the isobars round the 
Azores High has moved slowly (in accordance with the gentle pressure 
gradient) over a long stretch of warm ocean. This air spreads north- 
ward as a wind from the south-west over a sea whose surface gradually 
becomes cooler. Hence the surlace layers of air tend to fall in tem- 
perature, and at the same time their relative humidity rises nearer the 
saturation point. Very little lifting of the surface air is now required 
to form cloud and hence the northern parts of the middle Atlantic 
are a region in which extensive low cloud is frequently found. 

But, if pressure is low in the neighbourhood of Iceland, the air 
starting as a north-wind and following die isobars from the direction 
of Greenland is lound spreading from westerly points across the middle 
North Adantic. This air is moving from a cool source; possibly some 
part can be attributed to cooling over the Greenland plateau, but 
probably much more is due to the neighbouring ice-covered coastal 
waters. It then moves across an open water surface whose temperature 
increases to the southward. By die time it comes up against the air 
that has come round the Azores High it differs markedly in character; 
it is still considerably colder than the sea over which it is moving, 
whereas the air from the flanks of die High is warmer. Two such air 
streams of different temperature, and, therefore different density, do not 
quickly mix ; the lighter warmer air flows over the colder and denser air. 

We may now adopt the meteorologists' language and call such 
streams respectively 'tropical' and 'polar'. In our latitudes the term 
tropical is conveniendy used for air whose temperature in die lower 



layers as a whole remains above that of the sea surface; while in polar 
air the surface layers are cooler than the sea surface over which they 
are moving. Considering for convenience of description, the North 
Atlantic, it will be observed that throughout the year surface pressures 
tend to rise a little in high latitudes towards the pole. On the edge 
of this region of slightly higher pressure air more or less follows the 
isobars but with an outward component of motion; expressed other- 
wise, north-easterly winds arc dominant in high latitudes. 

At first it might appear that the north-easterly stream of polar 
origin, and the south-westerly stream from the Azores High might flow 
parallel to each other in opposite directions on either side of a well- 
defined boundary. In practice, however, such a system never persists. 
If we plot very carefully the observations of pressure, temperature, 
cloud and weather over such a boundary we soon find that conditions 
at the boundary are evidently unstable. Small masses of air may rise 
on very slight provocation and evidence of the process will be seen on 
the map as a very small area of slightly lower pressure at the boundary 
of the currents. But as soon as this occurs the air on cither side of the 
boundary which we may now call the 'polar front' deviates slighdy to 
follow the isobars round the incipient low. Such a movement leads at 
once to the stage shown in the third diagram below. Warm moist air 
rides forward and over the colder surface air north of the boundary. 

West of the centre, a tongue of cold air pushes southward, and in 
the form of a nose undercuts some of the warmer air at the surface. 
Near the tip of this advancing nose of cold air the moist warm air from 
the south is rapidly elevated; a belt of clouds and showers is char- 
acteristic of this area. If observations arc plotted for some hours the 
air is soon seen to be roughly following the isobars round the centre 
of a well-defined area of lower pressure which may after a day or two 
be upwards of a hundred miles in diameter. 

East or north-cast of the centre moist ' tropical ' air from the south 
is ascending steadily over the colder surface air beneath; accordingly 
a wide area of heavy low cloud develops in which the droplets are 
continually being formed and enlarged. Hence continuous pre- 
cipitation occurs in what we frequently describe as 'the rainy sector 
of a depression'. 

Once initiated and developed in this manner, the whole system 
lends to move more or less eastward along the boundary formed by 
the two types of air. Invariably, however, the protruding tongue of 


cold air on the south-western flank advances relatively rapidly and 
after some days careful plotting of die observations reveals that il has 
overtaken the line which we call the warm front; bounding the warm 

Fig. 1 1 

Meeting of Polar and Tropical air masses (based on a diagram by 

I'cttersscn, by permission of Messrs. McGraw-Hill) 

air and the rainy sector. The whole of the advancing warm air supply 
has thus been undercut. When this process has occurred the depression 
is said to have been 'occluded', and a sharp difference of temperature 
is no longer to be found ai the surface. The warm air above, however, 
is still as a rule undergoing uplift, and hence cooling with further 




FlO. 19 

o. f-4. Stages in the growth of a 
depression. 5. Occluded depres- 
sion. Note that on cither side 01 
the occluded front the surface air 
will be at much the same tem- 

b. Vertical section : 1 . Warm front. 
2. Cold front. 3. Occluded front 

condensation; cloud and more or less rain therefore still occur over 
a wide area. 

It will thus be recognised that 'frontal depressions' can be expected 
to develop wherever the necessary conditions arc found, namely, two 
markedly different streams of air moving at different speeds on either 
side of a well-defined boundary. The boundary between cold and 
warm air masses in the North Atlantic is found repeatedly in the 
neighbourhood of Newfoundland — Iceland — N. Norway, as we might 
well surmise from the geographical conditions in that latitude. The 
great majority of the depressions which affect us originate in this 
region and their centres travel from S.W.-N.E. along tracks to the 
north of Scotland. As each low passes the barometer falls and rises so 
ihat throughout the year there are frequent and considerable fluctu- 
ations; these in general are much greater in winter. Winter is the 



season of deeper depressions, steeper pressure gradients from margin 
to centre, and stronger winds. The energy of the depressions is no 
doubt associated with the extent of the contrast between the two air 
masses. In winter the seas to the south of Iceland are still open, and 
relatively warm; in January the average surface temperature is still 
50 off N.W. Ireland, and 40 on the S. coast of Iceland; not more 
than io° cooler than in July. But a little to the northward, for example 
in East Greenland in 70 N., coastal stations show that the air in 
January is on the average as much as 40° colder than July. This gives 
an adequate reminder of the greater intensity of winter contrasts 
across the polar front. 

It will now be evident to the reader that the Icelandic Low shown 
on the atlas maps of average pressure for January is not an indication 
of permanent low pressure diroughout the month. Such a map is 
rather to be interpreted as a demonstration that the centres of the 
deepest depressions are most often to be found passing over that 
region; on any given day, however, it is quite possible that for a brief 
interval after the passage of a low, pressure over Iceland may be quite 
high until the approach of a new low is heralded by a new fall of 
pressure. Further, the tracks followed by our Atlantic lows vary very 
much; their speed of movement varies too. It is broadly true that the 
majority of lows by the time they have reached the British coasts are 
occluded, and that occluded lows move more slowly and rather 
erratically. Tliis incidentally provides additional difficulties for all who 
would forecast British weather. Depressions, once occluded, begin to 
fill up more or less rapidly. 

The average life of a well-developed North Atlantic Low travelling 
from near south Greenland to the north coast of Norway is of the order 
of 5-7 days. The speed of movement, however, varies considerably, 
and while the great majority move more or less eastward, great 
variations in detail are observed. The diameter of the area affected 
by the circulation in a fully-developed depression is commonly of the 
order of 500-800 miles, but may range from less than 1 00 to as much 
as 2,000 miles on occasion. For a well-developed anticyclone 1,000 to 
1,500 miles is normal in our latitudes. 

Many depressions also approach us on more southerly tracks; for 
example, their centres may move along the south coast of Ireland and 
up the English Channel. Frequently such lows develop as secondaries 
to a primary centred north of Scotland; for just as the primary 



developed on the boundary between tropical and polar air south of 
Iceland, when the nose of polar air advances southward, the cold 
front at its tip is equally a line across which a marked contrast of 
temperature exists and it is very common for additional lows to be 
initiated upon it, and to approach us from a more southerly part of 
the Atlantic. It is also very important to recognise that contrasting 

types of air sufficient to engender 
active fronts may arrive from the 
neighbouring continent, especially 
in winter when the oceanic air is 
commonly decidedly warmer than 
that lying over Central Europe. 
Even on our Atlantic coasts a dis- 
tinction can frcquendy be observed 
between polar air which has had a 
long fetch over the ocean, and that 
which has arrived by a shorter 
route. For this reason it is commonly 
found that several minor fronts, each 
accompanied by a belt of cloud and 
more or less precipitation, are to be 
found over the eastern Atlantic, 
marking the boundary between polar air-streams with a longer or 
shorter track over the ocean waters. 

Much of our rain in Britain falls in association with the passage of 
occlusions. As wc have seen the passage of an occlusion is not marked 
by any sharp change of temperature. In theory, as the diagrams show, 
the air mass present at the surface is the same throughout. 

But in practice it is generally found that the polar air which has 
come all the way round the low does differ in temperature a Iitdc from 
that which it is now overtaking. In winter for example such cool air 
may have travelled from Scotland across a wide stretch of the Adantic 
before it swings back across England and the North Sea; when that 
has happened it is often a degree or two warmer than at the start. 
Similarly, in early summer there are times when the surface layers of 
the air crossing a wide stretch of the cool sea may be cooled a litde 
compared with the temperature at the start. 

Hence we find that there is often a small rise or fail of temperature 
as an occlusion passes, and hence 'warm' and 'cold' occlusions may be 

Fig. 13 

Charactcrisiir tracks of centres of 



distinguished. This is noteworthy because the belt of rain associated 
with a cold occlusion is generally narrower than with the warm type. 
But when wc consider what complicated tracks the air can follow 
among our islands, for example the effects of the varying breadth of 
the Irish Sea on the temperature of the air carrying it, it is clear that 
occlusions cannot always be easily labelled. The prediction of the 
precise length of time during which the rain is likely to continue is 
therefore not always practicable as yet 

It is important to remember that as the passage of occlusions is 
often an indication that a depression is coming to a stop or filling up, 
winds are in general nothing like so strong as with an active cold or 
warm front. Slow-moving occlusions in summer, in which a great deal 
of very moist warm air has entered above, are often productive of 
several hours of nearly windless rain. The early days of August 1948 
were marked by the passage of several such occluded fronts over 
Southern and Central England, and many places accordingly received 
more than the normal expectation of rain for the whole month within 
the first week. 

Persistent heavy rain with little wind is associated in the minds of 
many British travellers with summer holidays on the Continent. 
From what has been said it will be clear that the slow-moving occlu- 
sions associated with the shallow lows of summer are equally likely 
to develop where humid air supplies have spread into Europe. The 
present writer remembers too well the horrid puddles of Bonn and the 
dreadfully persistent rain at Salzburg, in an August when more 
Mediterranean air than usual had spread into the Tyrol, to be under 
any illusions that the weather is always better abroad. And the dis- 
comfort of twelve hours of remorseless August downpour at Boston, 
Massachusetts, with the temperature around seventy degrees remains 
as an even more disagreeable recollection. 

Lastly, even our cold fronts in Britain have rarely that sharpness 
of definition that we might in theory expect. One must imagine the 
cold air mass advancing over the land not in the manner of an evenly 
extended line of troops, but rather as a scries of platoons in echelon. 
As such a front advances on a winter afternoon, two or three well- 
developed squalls with their characteristic clouds and showers may 
pass over at half-hourly intervals. Between the onset of the first and 
the tail of the last temperatures may fall by as much as 8° or io°, 
in three steps of about 3 . The 'echelon' arrangement can be attributed 


to the varying frictional hold-up of the advancing cold air over hilly 
areas and plains, and also to such things as the local variations in sea 
temperatures round our coasts. 

For these reasons, too, the occurrence of lightning and thunder in 
association with cold fronts tends to be somewhat sporadic. Cold 
fronts arc best seen advancing over die sea or over open plains, for 
example that of West Lancashire in westerly weather, or over the 
Fenland if the wind is more northerly. Elsewhere, as the metcorologisi 
would say, they are apt to be somewhat diffuse, and rather rarely do 
they give the dramatic violence of those of the Great Plains of North 




* 8 " — -> 


1 V 


* 40 

52 - 


Fio. 14 

Sea temperatures round Britain 

left. Average isotherms (F°) for February, right. Average 

isotherms lor August 

Characteristics of the Air Masses reaching 
the British Isles 

Now that we have reviewed the workings of our great Atlantic 
depressions it will be clear that, so long as dicir eastward route of 
travel lies to the north of our islands, we shall experience winds from 
between southerly and westerly points; occasionally a few hours of 
northerly wind may be recorded, for example if the centre of a Low 
moves southward from the Norwegian coast into the North Sea. If a 
depression is not occluded we ought then in theory to experience 



during the day or two days of its passage, a spell of relatively warm 
air of tropical origin, making its way towards us from the region of 
the Azores; then a well-marked shower as the cold front passes, and a 
veer of wind from S.W. to W. with a sharp fall of temperature as the 
polar air arrives. 

But in practice, we find that the centres of some depressions moving 
on more southerly tracks pass south of England, giving us easterly or 
north-easterly winds which may originate over die continent, or may 
be derived, tor example, from the northern Norwegian Sea. Some- 
times, instead ol" receiving our polar air by a rather long sea route over 
the Atlantic, we may get it directly from the Arctic. Air coming from 
the edge of the Arctic pack-ice near Jan Mayen crosses less than 900 
miles of open sea to Scotland: whereas if it reaches us from the region 
of South Greenland it may have crossed anything from 2,000 to 3,000 
miles of ocean. 

According to the source-region and the route of travel, therefore, 
we may classify the six 
principal types of air 
reaching Britain. Most of 
the air reaching Britain 
can be traced dirccUy from 
one of three source- 
regions; regions, that is, 
in which the air has been 
relatively quiet for some 
days and has had time to 
acquire considerable 
homogeneity of tempera- 
ture and humidity 
throughout its mass over 
a wide area. Sufficiently 
quiet regions for this pur- 
pose are found in the 
permanent Azores high- 
pressure region, and over 
the high Arctic latitudes of 
the Polar Basin; in winter 
another very well-marked 
source, this time of 




— I — I — I — i-»- 

Fio. 15 

Generalised directions of movement of ihe 
characteristic air masses reaching Britain 



extremely cold surface an, is found from time to time in anticyclones 
centred over Northern Russia, representing a north-westward displace- 
ment of the continental anticyclone mentioned earlier in this chapter. 
In summer too, conditions are often quiet enough on the continent for 

the air to acquire during two 
or three days' travel well- 
marked characteristics o/ 
warmth, and hence dryness. 
The six types of air are 
known as Maritime and Con- 
tinental Arctic ; Maritime and 
Continental Polar; Maritime 
and Continental Tropical. 
These titles are often abbre- 
viated as mA, cA; mP, cP; 
ml' and cT. It may be added 
that in low latitudes 'equa- 
torial' air is distinguished 
from 'tropical' air, but that 
equatorial air never reaches 
our latitudes. Our tropical 
air can be traced as a rule to 
the region between 25 and 
40 N. As we have already 
seen, its principal character- 
istic is that over the North 
Atlantic it is warmer than 
the sea surface over which it 
is moving. Several phenom- 
enally warm nights late in 
November 1947 were due to 
the arrival of a tropical air- 
stream from as far south as 
20 N. in the central Atlantic. 

Fio. 16 

1300 nrs. 5 November 1938. Exceptional 
warmth. Maritime tropical air with low 
cloud and fog on S.W. coasts, subsiding east 
of Wales and giving clear skies (sec p. 5) 

As we have seen, the modification of air masses on their way 
towards our shores takes place largely in the surface layers. The 
properties of air masses are much more conservative at high levels. 
For example, maritime polar air leaving the Greenland seas may have 
a surface temperature of io° and at ten thousand feet -20°F. Arriving 


off" the south-west coast of Ireland in January when the sea temperature 
is still about 48 , it blows over the land with a temperature of 40 or 
so in its surface layers; but at ten thousand feet the temperature may 
sdll be -10 . We thus see how necessary it is for the forecaster to 
have his upper air data at hand. Moreover, the properties of the air 
masses reaching us play an integral part in the impression we obtain 
of the country; visibility, the intensity of light, the type of cloud and 
its extent are all at once affected. 



™ /// 

,980 y / 
■ 9S * / / 



yT~ ^%\J- 

.988 / 

J)V®59 --^ 

992 / 


"icW 6 J^ / ' 

■§,$5b / 


y^3 jj[655 ^J 


1000 jS 


EfS9 -J 

S JW* 

9 ȣmf 




t / 

> 59 

VJCN - — "'032 


Maritime Tropical Ait 
fulfils the conditions des- 
cribed earlier; flowing as 
a humid current over a 
colder surface, its relative 
humidity by the time it 
reaches Britain is high. 
If its motion is exception- 
ally slight it may give 
surface fog over our 
south-western approach- 
es and on coasts such as 
those of S. Cornwall, S. 
Pembroke, and Anglesey, 
especially if on its course 
it has been warmed 
slightly over N. France 
at the season when the 
sea Is relatively cold, 
namely in late spring or 
early summer. In general, however, it is moving fairly briskly with 
appreciable turbulence over the sea; hence the saturated air is found a 
few hundred feet above the surface in the form of very extensive low 
stratus or strato-cumulus cloud. This is particularly noticeable in winter 
when stronger winds prevail and the air, being cooler, requires less water- 
vapour for saturation. Hence, the cloud-base of the characteristic 
St. and Sc. tends to be lower in winter than summer. Above the 
low cloud over the sea skies are often clear; and the surface air is 
stable, being slightly cooler than that above the cloud. mT air moving 
inland in summer is moving on to a land surface which in the daytime 

Fio. 17 

Synoptic chart of maritime tropical air over 
England. i8oohrs.,at November 1947 (sec p. 5) 





is receiving some radiation, even through the cloud. Hence the surface 
layers become warmer; cloudbasc, therefore, tends to rise to a higher 
level. As we have seen, however, the skies above the cloud are generally 
clear and the air relatively dry. It is, therefore, not uncommon to 
find that in summer the cloud in mJ air moving inland becomes 
thinner, and in the daytime breaks up into detached cumulus; for as 
the sky clears the ground warms sufficiently to give the needful rising 
currents in moist air. 

In winter on the other hand, mT air has already become cloudy 
as it approaches our coasts, and moving over the land which is in 
general colder than the sea, the low cloud persists and thickens. 

Indeed mist or even fog may 
prevail at ground level espec- 
ially if the ground has been 
previously chilled by over- 
night radiation, or for exam- 
ple, by recently melted snow. 
English readers will recall the 
many milder days of winter 
when a light south-west 
wind, a dull grey sky, and 
humid surface air prevail all 
over the Midlands: visibility 
is rarely more than two miles 
or so. In the nearly saturated 
air under the grey lid above, 
smoke from the great cities 
drifts slowly over the coun- 
try, so that large areas for 
many miles to leeward of the 
cities record even poorer visi- 
bility and greater dullness. 
Temperature on such days 
on a December afternoon 
hovers about 50 on the 
Cornish coast, while 46°-47° 
Img. 18 inland is characteristic. In 

Synoptic chart of Continental Tropical air Juty on & e Other hand, after- 
over S.E. England. 1200 hra., 3 June 1947 noon temperatures may he 

in the region of 66°-68° on the Cornish coast, but with clearing skies 
the Midlands have a humid relaxing afternoon, only too well known 
at Oxford and Cambridge, with a maximum of the order of 74 and 
intermittent sunny intervals between widespread lumpy cumulus with 
summits ol varying height. 

Continental Tropical Air. By contrast this is warm and dry; but even 
in the south-east of England it arrives rather infrequendy in summer, 
practically never in winter. Sometimes in summer pressure is relatively 
high over the continent, while a shallow depression approaches slowly 
from the Atlantic. South to south-easterly winds bring air which may 
have been warmed for some 
days over S.E. Europe; the 
dry clear air travels across the 
well-warmed continent and 
is capable of giving on rare 
occasions very high temp- 
eratures indeed in N. France 
and S.E. England, the Chan- 
nel not being wide enough 
here to cflect more than a 
narrow surface stratum of 
the air. If the stream of air 
maintains its course north- 
ward over the land, the 
temperatures may attain 
high levels in Central Scot- 
land, but more rarely and 
for an even briefer period 
than in the south as the 
adjacent map suggests. 

In winter it is very rare 
indeed to receive air which 
is both warm and dry from 
the south. Even North Africa 
at that season is relatively 
cold. Rccendy the phenom- 
enal records of 70 on Nov- 
ember 5, 1938 and 75° on 

Fio. 19 
1200 hrs.. 29 July 1948. Continental 
Tropical air; great heat in Manchester 


March g, 1948 give rare hints of the possibilities of winter warmth but 
it is doubtful whether the origin of the air on these occasions really 
lay so far south as N. Africa. (Cf. the charts shown on pp. 72 and 88). 
It will be evident that in summer the approach of the above- 
mentioned Atlantic depression widi its cooler maritime air from the 

F10. ao 

a November 1930. Rain ahead ol advancing warm front A; 

heavy thunderstorm in London at 10.30 a.m. with severe 

squalls at the passage ol cold front B (noted by F. H. Dight, 

Q. J. Roy. Met. &, 1 93 1) (Notation, p. 5) 

S.W. usually breaks up any brief hot spell; this will be discussed again 
later when we consider the great thunderstorms which usually ter- 
minate hot weather. 


Maritime Polar Air. In its many varieties this is without doubt the 
most common type of air we receive. Let us look into its properties. 
In winter cold air is approaching over a warmer sea. The surface 
layers therefore become warmer, and ai the same time pick up mois- 
ture; while above, the air is still relatively cold and dry. At die surface, 
air coming from S. Greenland can frequendy be expected to increase 
by 30 in temperature in January; but at 10,000 feet the increase is 
only about io°. Hence as the air travels towards us the lapse-rate 
increases to a point where the surface air is generally unstable; and 
the intermittent ascent of moist air from the surface gives extensive 
belts of cumulus over the sea. These are well seen in winter on our 
westward coasts almost whenever there is a strong westerly wind, and 
the hard-edged cumulus gives characteristic stormy but vivid sunsets. 

Sometimes when the air is exceptionally cold above, the towering 
clouds become cumulo-nimbus and thunder and lightning occur, 
especially when the wind blows strongly against coasts which give a 
slight extra impetus to the rising currents, for example, the West 
Highlands, and even the cliffs of Cornwall and Sussex. In all these 
regions winter thunderstorms are reported with some frequency, 
compared with the Midlands. Occasionally a well-marked cold front 
causes thunderstorms to travel right across England even in a winter 
month. The weather map above shows a depression followed by two 
cold fronts, the first of which gave a prolonged thunderstorm lasting 
nearly an hour in London from 10.30 onward on the November day 
in question; it wasdiscussed by Mr. F. H. Dight (Q,.J. Roy. Met. S., 1931). 

The degree of instability in mP air depends a great deal on the 
length and route of its travel over the warmer ocean. Given a very- 
large low west of Ireland the polar air coming round it may have 
travelled far to the south, returning towards us as a wind from the 
south-west. Under these circumstances it has travelled some distance 
over a sea warmer than our own waters, and thus as the surface layers 
approach us they become more stable and, like tropical air, may give 
considerable low stratus cloud and very similar temperatures, both in 
summer and winter. In such returning polar air in summer, however, 
the upper air is still quite cold. The low strato-cumulus breaks up 
over the warmer land (in a similar fashion to mT air above) and the 
rising currents give cumulus. In mT, however, the upper air was 
warm and dry; but with returning mP it is considerably colder. Hence 
once convectional cumulus has been started it may ascend to high 


levels through the cold environment above, and such rapid growth 
may result in inland showers, sometimes accompanied by thunder. 

Maritime Arctic Air may conveniently be considered here; it is 
merely an exaggerated form of polar air with a relatively short sea 
travel. In winter it feels bitterly cold, as normally it reaches us as a 
strong north wind, often attaining gale force on exposed coasts. Such 

cold air flowing over a 
warm sea is extremely 
unstable, and gives char- 
acteristic sharp showers 
generally of snow and hail 
in the colder months. 
These developing over the 
sea, become especially 
noticeable in many coastal 
regions of Britain, such as 
those of East Kent, North 
Norfolk and Lindsey, 
Holdcrness, Cleveland and 
Durham, Banff, Moray, 
Nairn, Caithness and 
Sutherland; also North 
Wales. In every case the 
unstable air from the north 
impinges on a sharply- 
rising coast; the additional 
ascent of the surface air 
gives vigorous showers. On such days of Arctic air the towering cumulo- 
nimbus clouds over North Norfolk can often be seen from Cambridge; 
those over Nordiumberland and Durham can be seen from the sunlit 
summits of the Lake District in the lee of Scotland. For only the surface 
layers of the air are moist; hence when they have descended on the lee 
side of a mounrainous region such as the Scottish Highlands they arc 
drier and much less likely to give cloud than where they have just 

Fig. 21 
2!! February 1937, 7I1. Maritime Arctic air 

Plate 9 

Oudimnoton. Northamptonshire: on a fine mild January afternoon. Temperature 

♦4 . Characteristic patches of strato-cumulus in mild south-westerly air stream 

round a High over N. France. (Cf. fig. 32, p. 98.) 





traversed a wide stretch of open sea. Hence with Arctic air the Lake 
District enjoys long spells of superbly clear sunshine, and the like is 
true of South Wales. 

Even in May, mA if it arrives is likely to give snow showers in all 
the districts named above. In a winter month such as February, on 
the north coasts of Scotland it may arrive with a surface temperature 
of 30 or even a little lower, and if it is blowing hard day-time maxima 
are scarcely likely to exceed 35 even in the south. Maritime Arctic 
air sometimes reaches us from the N.E. across Scandinavia, having 
come originally across the open sea from the north. 

Continental Polar and Arctic Air may similarly be considered together. 
Both reach us from the south-east, east or north-cast, in association 
with high pressure over Scandinavia and the Baltic region. In summer, 
cP in its surface layers becomes quite warm as it travels across Central 
Europe, and it reaches us as a rather hazy easterly wind cool on the 
North Sea coasts but warm inland. The origin of the air can be traced 
to Central Russia. In winter, cold, dry and frequendy cloudy weather 
prevails; the cloud being largely the result of the passage of the air 
across the North Sea. While the North Sea is still fairly warm, as in 
late autumn, the instability resulting when cold air Hows over the 
warmer sea may give rise to showers near the east coast. 

Continental Arctic Air derived from Northern Russia or even N.W. 
Siberia gives us our most severe winter weather. If the continental 
high of winter extends, generally in the form of a separate cell with a 
centre lying across North Russia to Scandinavia, a large area of 
intensely cold stagnant air is still further cooled to a considerable depth 
as a result of radiation over the snow, and makes its way outward 
across Germany and N. France to England; if it reaches Scotland, it 
generally does so from the south or soulh-east rather than from due 
east. Air of this type crosses the Russian frontiers with a temperature 
which is sometimes below zero; crossing Germany its temperature 
slowly rises and it becomes drier, largely as a result of mixing with 
warmer air ahead and above. But even in Holland and Eastern 

Plate 10 

Little Eaton, Derbyshire: birch wood with hoarfrost. February. Very character- 
istic of quiet winter days when temperature has just fallen to the freezing point 

at night. 
C.B.S. G 






Francc the thermometer may still read below io°F. (see chart below) 
with a strong cast wind to add to the discomfort; for towards the 
margin of the anticyclone the pressure gradient is steeper, and under 
such conditions a gale is often experienced down Channel. The efTect 
of the sea, however, is very apparent on the English coasts ; temper- 
ature and humidity rise considerably and with the addition of moisture 

in the turbulent surface 
air a good deal of low 
stratus cloud forms fre- 
quently covering almost 
die whole sky ; cloud base 
lies between 1,500 and 
3,000 feet. In December 
1938 a severe outbreak 
of this type of air gave a 
maximum temperature, 
on die 2 1st, of only 22 
at Lympne in Kent, 25° 
at Brighton, 26 in Lon- 
don. But on the same 
day maxima on the 
adjacent condncnt were 
far lower, io° at Calais 
for example. Perhaps the 
most appalling outburst 
of such condidons we 
know of occurred at the 
end of December (O.S.) 
1739. There is reason 
to believe that with an 
easterly gale blowing the temperature in London was below 15 on 
that occasion; in Holland the gale was accompanied by temperatures 
from -2 to -f-2°F. for many hours. 

The warming effect of the sea is very evident. Even the coldest 
east wind down-channel rises to temperatures near the freezing- 
point, and on the coasts of North-East England and E. Scotland 
the same slight rise is evident. At the same time however the surface- 
layers become more unstable and a severe outbreak of air of this type 
frequently gives rise to snow showers on and near these coasts, although 


Fio. aa 
ao December 1938. Continental Arctic air. 
An exceptionally cold outburst; warming over 
sea gives instability snow showers in E. Eng- 
land, frost at Scilly Isles and S.W. Irish coast 



(Hl«fc »<»* ,oz 5 . 

they are more conspicuous as a rule in die maritime-Arctic air pre- 
viously described. 

Continental polar air in winter differs merely in degree from that 
derived from more northerly sources. The normal keen south-east 
to cast wind with a temp- 
erature just above the 
freezing-point at inland 
stations and a grey sky is 
more often than not 
derived from the great 
plains of S. Russia, 
associated with high pres- 
sure over Germany. The 
course followed by the air 
depends largely on the 
position of the region of 
highest pressure. (Cf. fig. 

«5. P- 7')- 

In describing the 

characteristics of the six 
principal types of air 
reaching tins country, 
we have considered their 
behaviour in high sum- 
mer and the dcpdi of 
winter. Little has been 
said with regard to spring and autumn. At these seasons there is still a 
well-developed high-pressure area in the Central North Adantic provid- 
ing a source region for our maritime tropical air. Towards the continent, 
however, 'the great Siberian High' is no longer found. In spring it is 
more common to find relatively high pressure over the region of the Baltic 
and Scandinavia, and sometimes over the northern Norwegian Sea. 
It will readily be seen that with persistent high pressure in these re- 
gions we may expect many days of winds from the north-east quadrant; 
and one of the most characteristic features of the British climate is the 
relatively high frequency of winds from that direction from March to 
May. Moreover, with regard to continental polar air in a month such as 
May, the surface layers become quite warm and dry as they cross Central 
Europe as the skies are generally clear; but the North Sea at that season 

Fio. 33 

Synoptic chart of Continental Arctic air; note 

increased cloud after crossing North Sea. 

0600 hrs., 8 February 1947 (Notation p. 5) 



is stiH cool. We find then that relatively warm air is approaching us over 
a cooler sea, producing as a rule an 'inversion-layer' at an altitude of the 
order of 2,500-3,000 feet. Below this layer, the turbulent surface flow 
often carries up enough moisture from the sea surface to form extensive 
cloud. Hence the frequent formation of the 'North Sea Cloud' — stratus 
or strato-cumulus — in dry weather in spring, and early summer, extend- 
ing for many miles inland from our east coasts, sometimes right across 
Southern England. In late autumn on the other hand, we have seen that 
the same pressure distribution still tends to give cloud in the air cross- 
ing the sea. But the sea is relatively warm while the land is cooling; 
hence the air coming from land to sea is not as a rule so stable as in 
May. Cloud is more disturbed, and showers arc more frequent. The 
effect of the North Sea on cloud formation has been carefully studied, 
and a summary has been given by Mr. E. Gold, formerly of the Meteor- 
ological Office in a lecture to the Royal Meteorological Society on 
Weather Forecasting. It will be found in the Society's journal for 1947. 

Maritime polar air tends to become unstable due to warming over 
the sea in winter, and over the land in summer. At the intermediate 
seasons the chance of cumulus development and of showers is if any- 
thing slightly less, and the relative dryness (on the average) of April 
and September is partly attributable to this. 

Our air-masses may not only be somewhat modified, depending 
on the breadth of the adjacent sea which must be traversed before they 
reach us; they also can be expected to behave slightly differently in 
the several months. At all times of year, however, the march of the 
Atlantic depressions towards Europe continues to play a major part 
in our weather. They and their secondaries draw in the air-masses 
we have described; diese in turn with their characteristic cloud and 
weather, give rise to the changeful effects of light and colour in our 
landscape which play so large a part in the beauty of these islands. 
The principal departures from the normal changefulness arise whenever 
the march of the depressions is either temporarily interrupted or 
diverted. We may now discuss why some years are characterised by 
considerable spells of settled weather. 

'Settled Weather': The Movement of Anticyclones 

An anticyclone is a region of higher barometric pressure surrounded 
by closed isobars, according to definition; and we have already used 


the term for the large high-pressure region which generally covers the 
central Adantic between the Azores and Bermuda, and for the 
Siberian High of winter. 

At the surface it is observed that the actual flow of air is slighdy 
outward across the isobars; this means that the great cushion of dense 
air which we describe as an anticyclone leaks. To compensate for this 
there must be some subsidence in the centre; as we have seen, this 
means compression and warming of the descending air, hence a de- 
creased relative humidity and cloudless skies. 

Smaller transient anticyclones occur from time to time, as detached 
masses of relatively quiet air; in general they are larger in area than 
depressions and move more sluggishly. 

Detached areas of subsiding warm air up to fifteen hundred miles 
in diameter frcquendy spread north-eastward from the Azores High. 
In winter, surface masses of cold dense air from the Arctic or from 
Siberia sometimes spread and rest for several days over Scandinavia 
and the Baltic. A large, slowly-moving High tends to hold up the 
circulation i.e. the movement of depressions. Yet such Highs have 
been described as themselves a necessary part of the circulation, 
inasmuch as considerable forced ascent of surface air occurs in depres- 
sions and this is compensated by the large scale subsidence in anti- 
cyclones. Highs indeed must exist; but exactly why in some years they 
remain faithful for weeks to some particular part of the North Adantic 
or N. Europe, while in other years they scarcely develop there at all, 
is one of the biggest puzzles our atmosphere has to offer. 

Occasionally a persistent anticyclone in an unwonted part of the 
Atlantic produces very marked departures from average conditions 
over several weeks. Examples can be cited in February 1932 and 
April 1938. In both months pressure remained persistently high to the 
west of the Hebrides. The normal Atlantic depressions repeatedly 
followed tracks far to the north beyond Iceland, or skirted the anti- 
cyclone to the southward; Britain in general was only affected by the 
movement of air on the margin of such lows. 

Both months were extremely dry. In February 1932 light northerly 
winds prevailed, but the month was not cold; for it will be evident that 
the surface air had either moved gendy round the High from the region 
of the North Adantic, or had subsided from higher levels; the north 
wind was not of Arctic or even Continental origin. With clear skies 
at night there was a little frost, but the days were relatively warm; 

8 4 



V net 

lOtfl > 

i : - - . \ 


io2i J\r 



V- 'ooe 



,f, -%~x. fSrT"-*: 

sj> ioo» 

« o* r ^y^CP^C^5 



/ ^* , M^^ir!7 

U.- ioio 

*o?9 — 7^/*l^^»v , 

X y /ft f"*\ 

•5in * '•" 

1030 jr&>'*>%Z'/ >, 

— v 

•0 J? -.'^^^ >< >^7\\ 


.053 —Sf , ' -V =?*r • 

^■\ ft ^^V '"^ 


? -v^r 55 ! r*^ ' 


/ / ^*-y 

/?•*'!/ ^ 


^^-—^" Jib 16 


•/ *"*^ 

A '/H 1- " 

1033^,' '.f* ' 

I0J3 >_ , >V 

•'$&< ^T 10 ) / 


-r^v/ L--riAI 


£,,- A w-y (•xS'J-iou 

V'Cii^* V/- -14 


-^- ^ — \ ' ry 1 If* ,0 ' 3 

A- -"v^T* y'V , I V-— 1-1014 

r ^^"^^ -^ / * 1 A 


y^^y. S J*-*^_j_ J z^'ZZ~j' m 

10 "- 'J^-f^^si-4- 

1031 ^ "^* ■ "" ,**""" ^ ^^ 


c:: -^ \ 

WJ* .. 

IOJS -i\-fl _ KIO 

m>4 \ X 

ms v\ 



^*|TT J -iififltirr ~ "-" 




r^N* / \ YV \ X 

1019 —J 

' /vTvA-Xa 

^ ' 

\"Ll d\ 1*4 \ 


1 lI'^vA/^ t=-r'015 



Fie. 34 
Three exceptional anli-cyclonie 
montlis; showing wind roses and 

mean pressure in mb. 
a. February 1934, b. February 
1932. c. April 1938. 

Normal pressure distribution 
shown by dashed lines 
(For notation, sec p. 5) 

rather cloudy with low stratus towards the east, but frequently sunny 
on the west side of the country, and so dry that at places in the wettest 
part of the Lake District no rain whatever was measured during the 
month. For the first three weeks of April 1938 conditions were very 

February 1947 was characterised by an extremely obstinate anti- 
cyclone towards the Norwegian Sea and North Greenland, that is, 


centred farther away from these islands. On its southern margin, 
very persistent easterly wind prevailed, completely reversing the 
normal circulation as the map shows (Fig. 25, p. 86). 

Among the most remarkable summer montlis dominated by a 
large and persistent anticyclone were June 1925, August 1947 (Fig. 25, 
p. 86), August 1955 and September 1959. In June 1925 no depression- 
centre approached within 250 miles of our coasts; much of Southern 
England was rainless. The skies were so cloudless, especially where 
the gentle flow of air had come over a long stretch of land, that places 
on the Cornish coast had an average of 12^ hours of bright sunshine 
daily, nearly 80% of the maximum that could be recorded. In August 
1947, the distribution of rain was even more remarkable; no rain 
whatever fell in Glasgow and over much of the W. Highlands, in 
exceptional contrast to the normal experience. 

Winter anticyclones in our maritime climate, however, frequenfly 
give extremely persistent cloudy and overcast skies. Even in a summer 
anticyclone the increased movement of air towards the outer margin 
may be enough to build up a good deal of low stratus or strato- 
cumulus where the air crosses a cool sea. In autumn and winter the 
establishment of a High following a spell of typical rainy weather 
means that the ground is damp, the skies clear, and the air quiet; all 
these condidons are extremely conducive to the formadon of extensive 
radiation fog. After a calm November or December night this is often 
sufficiently thick to last through the next day, at the season when the 
sun's rays have little power to penetrate the fog and warm up the 
ground beneath. Meanwhile the air above the fog is commonly 
subsiding and becoming warm; hence the inversion-layer in which the 
fog exists becomes more marked. Any movement of air within die fog 
is not likely to lift it through the inversion; instead, the fog tends to 
drift along the surface. In this manner we frequendy find that radi- 
ation fog formed over the land in quiet weather drifts gendy over the 
adjacent estuaries or narrow seas. The mouth of the Thames and 
nearby Channel is thus liable to be foggy in cold winter weather; the 
same applies at the mouth of the Mersey. 

An extensive anticyclone in which fog has formed generally tends 
to move eastward in winter and to merge with the high pressure area 
towards Central Europe. Hence the winds on its western flank gradu- 
ally become southerly and increase in force. Fog is then lifted off the 
ground, exactly as with the advecrion fog over the sea which was 


described in the last chapter; but such fog only becomes low stratus. 
It cannot ascend into the warmer air above, and so it is liable to per- 
sist for as long as the anticyclone remains in our neighbourhood. Very 
persistent cloud of this type prevailed during the cold February of 
1942 over Southern England, and even more during the severe 
December of 1890. This still ranks as the most sunless month on record 
for most of England, and gave rise to the phrase 'anticyclonic gloom' 
frequendy used by British meteorological writers during the suc- 
ceeding decades. 

From March onward to October anticyclones may give consider- 
able low cloud on their flanks while the air has crossed the cooler seas 
as we have already seen; but large areas are free from cloud, local 
radiation fog is easily dissipated early in the day by the more powerful 
sun, and high sunshine durations prevail especially in the lee of 
mountains. April 1938 and June 1925 have already been cited. 

Nevertheless, the extent to which we are affected by transient 
highs varies very greatly from year to year, and meteorologists have 


».. ,c !' V' 

•on '/7>2= 

K.7 ><r '*** 

"'VM1 • 

rC-- - x>o* 

10,3 ?c«^ 

— ~* .''^ 


~-^Z\ s -"»o 

■on . THSfire- 


l0,0 0^fcJv>* 

^~^-i r~ ion 


>^.- .iy£>» ■ 

l00 *ll_^ •'V.' v*^ 



—C^- J \.'' 5**^ 

ioot _?isr ^^t 7^ 



tOOfi V _ ^"Ni 7 

V ^J-^^ / st!] 



ioo5^rC "^^_ _^< 

V^, «* — 

/V^.** "^J* 


Nv >v A 

1004^ ,- *5C, 

Tji,- ' ".J. 


"■ ''•' -* 

;^^C s 


J ^-— . , 


Fic. 35 

Wind directions and mean pressure shown in millibars for 

February and August 1947. Normal shown by dashed lines 

not yet fully explained their behaviour. Somcdmes a large and healthy- 
looking anticyclone disappears unaccountably in a day or two as in 
mid-September, 1932. In occasional years a succession of anticyclones 



drifts over us throughout the summer apparendy as detached portions 
of the Azores High. Moving slowly from west to east they fend off the 
Adantic depressions for long periods, and Southern England experi- 
ences a very fine dry summer like that of 192 1, or to a less extent 1933 
and 1949. Out on the northern margin of the Highs, however, Scotland 
comes in for the stronger westerly winds associated with the passing 
depressions far to the north; in such years the Scotrish summer is 
relatively cloudy and much cooler than that in the south. 

Around the margins of our anticyclones there is a tendency for 
the surface air-streams to be slightly more convergent on the N.W. 
flank and on the S.E., with a corresponding tendency for slight 
divergence on the N.E. and S.W. flanks. Hence when a persistent 
anticyclone hovers to the north-westward, as in the dull cold May of 
1923 and July-August 1931, these tendencies for cloud development 
particularly affect South-east England, especially as the air moving 
round the High has a long "fetch "down the North Sea. The com- 
parison between Scodand and London is then reversed. Southern 
England especially is affected by the low stratus from the North Sea, 
and also by the cloud on the outer margins of depressions which find 
their way eastward in such seasons across N. France. Scotland, and 
notably Western Scodand in the lee of the mountains, has clear skies 
and sunshine. The report of the fine summer of 193 1 in the Hebrides 
came in a year when the South Coast summer was parricularly sunless 
and there was little money for holidays abroad, and undoubtedly the 
result was to attract many Londoners to Scodand in the next year. 
August 1945 and late Junc-carly July 1948 gave for two weeks a 
somewhat similar experience. In August 1947, however, the High 
lay a little farther east and was more intense; clear skies predominated 
everywhere. In the summer of 1958 the situation was again like 193 1. 

In general, however, summer anticyclones more frequently spread 
across South Britain than elsewhere; they are accompanied by wide- 
spread small fine-weather cumulus developing inland in the afternoon, 
especially to the northward, in the light westerly current overlain by 
warmer subsiding air. It is rather in spring that we most often find 
anticyclones centred to the northward. One result is that in May the 
island of Tiree in the Southern Hebrides experiences on the average 
more sunshine than almost any other part of the British Isles. 

The air moving gently round an anticyclone may fall by virtue of 
its origin in any of the classes already mentioned. But as it is moving 



slowly it has more time to become modified. Expressed otherwise, 
throughout the warmer months the sun has more opportunity to warm 
the land, and to modify whatever type of air is crossing the country 
in the direction of greater warmth and dryness. Hence with a normal 
summer anticyclone centred to the southward and a westerly wind, 
Eastern England has higher temperatures than the west. Rather 
rarely in winter when there is already a snow-cover, an outburst of 
severely cold continental-Arctic air is followed by calm, clear skies 
consequent on the rapid build-up of pressure over Britain. In such a 

case the surface air, already 
cold, falls very rapidly in 
temperature at night over 
the snow; and the occurr- 
ences of minima below zero 
in England, at least, can 
almost all be attributed to 
this sequence of events. In 
January 1 940, -6° was record- 
ed at Bodiam in Sussex, and 
-4 in Kent (at Canterbury) ; 
a century earlier, January 
1 838 gave minimum tempera- 
tures which by modern stand- 
ards undoubtedly indicate 
more than 10° below zero 
in similar inland locations. 
Phenomenal dry warmth, 
such as the maximum of 77 
recorded at Wakefield in 
March 1929, is again a con- 
comitant of a gently moving 
stream of dry warm air from 
the south combined with 
clear skies; even more 
remarkable was the maxi- 
mum of 75 in the London 
suburbs as early as the 9th 
in 1948, with 74 at Cromer 
and 70 as far north as Hull. 

Fig. 26 

Noon, 9 March 1948. Maximum in Cam- 
bridge 73 at 1500 bra.; compare situation 
lor 5 November 1938, p. 72, and 18 Jan- 
uary 1947, p. 98 (Notation, p.5) 


Incidentally the reasons for the lack of a sea-breeze at Cromer on this 
occasion in spite of the proximity of the North Sea at its coolest, are 
noted on p. 141. On 12th March 1957, 72 at Elgin is also noteworthy. 


Fio. 27 

Mean daily pressure at Kcw, 1881-1940: frequency of 'anticyclonic' 

and 'stormy' days, British Isles, 1889-1940 (by courtesy of Dr. C. E. P. 

Brooks and "Weather") 



In any discussion of the atmospheric circulation over the British 
Isles mention should be made of the extent to which spells of a par- 
ticular type of weather tend repeatedly to occur at the same time of 
year. Folklore of all countries includes beliefs suggesting a regular 
tendency for abnormalities at certain seasons. In 1869 Alexander 
Buchan wrote his famous paper on Interruptions in the regular rise and fall 
of temperature in the course of the year, and much work has since been done 
in this direction. The reality of Buchan's six 'cold' and three 'warm' 
spells may be debatable as he based his findings on a limited number 
of years of Scottish records. But considerable evidence was collected 
during the war in the Meteorological Office regarding the recurrence 
of such 'singularities', as they have been called, in North-Wcst Europe 
and an account of the results has now been given by Dr. C. E. P. 
Brooks, under whose superintendence many important climalological 
investigations have been made. • 

F10. a8 

The annual march or mean daily temperature at Tolland Bay, 1886-1918 

By permission, from Q.-J- Roy. Met. S., 1919. Cf. fig. 44, p. 147 



Dr. Brooks' paper clearly shows (fig. 27, p. 89) that in South- 
East England the chance of disturbed weather reaches a peak in 
December from an early summer minimum; but there is also con- 
siderable variation in the chance of anticyclonic days in different 
months. Early June and the second week of September arc well 
favoured in this respect. Some minor tendencies arc also evident in 
many years, for example stormincss in late November and early 
December, and again in the past 52 years disturbed weather has 
supervened sufficiently frequently during the week after Christmas to 
make another peak in the record. The tendency for cold anticyclonic 
weather towards the middle of February has been recognised by others 
(E. L. Hawke in Buchan's Days), and it appears in the annual course of 
the Totland Bay temperatures (p. 90). But it is interesting to note 
that Buchan's third cold spell associated on the Continent with die 
'Ice Saints of May' (1 1— 13 May) does not appear to occur with any 
consistency according to this later investigation. 'St. Lukes Summer', 
a brief anticyclonic spell in mid-October, occurs with fair frequency. 
It will be interesting to sec how many of these singularities are accepted 
after we have acquired a further fifty years of observations; they are 
indeed interesting but several of them can scarcely be regarded as 
regular enough, as yet, to satisfy would-be holiday-makers. Further 
reference with regard to such spells can be made to a paper by 
H. H. Lamb. 

A rather different approach to this problem, restricted to the 
evidence of abnormally high and low temperatures for the season, has 
been made in the Quarterly Journal of the Royal Meteorological Society by 
E. L. Hawke for the Greenwich temperatures (1841-1936) and by 
W. Dunbar for the Kilmarnock temperatures (1902-41); the tendency 
for occurrence of extreme values is by no means the same at the two 
stations and the incidence of extreme temperatures of Kilmarnock 
does not in this century appear to support the validity of the periods 
which Buchan originally derived for South Scotland. 

The search for periodicities of one sort or another in meteor- 
ological events will undoubtedly continue to fascinate large numbers 
of inquirers. At the present time however, it must be emphasised that 
any claims that such tendencies exist must be treated with the utmost 
caution. At the very least we must have longer terms of records, kept 
with sufficient strictness to stand up to rigorous analysis. 



Chapter 4 

References to frontal phenomena and frontal behaviour abound in text- 
books and journals. 

The Daily Weather Report of the Meteorological office may be watched 
with profit. 
Admiralty Weather Manual. London, H.M.S.O., latest ed. 
Bartholomew's Atlas of Meteorology (1899). Edinburgh, old, but 

useful for purposes of general illustration as far as the British Isles 

arc concerned. 
Belasco, J. E. (1945). Temperature Characteristics of Different Air 

Masses over the British Isles. Winter. Q,.J. Roy. Met. S. 7/: 351-76. 

(1948). Incidence of Anticyclonic Days and Spells over the British 

Isles: Weather, 3: 233-42. 
Brooks, C. E. P. and Mirrlees, S. T. A. (1930). Irregularities in the 

annual variation of die temperatures of London. Q,. J. Roy. Mel. 

S. & 375-88. 
Brooks, C. E. P. (1932). The origin of anticyclones. Q,. J. Roy. Met. 

S. 5 8: 379-88. 

(1946). Annual recurrences of weather: 'Singularities'. Weather, 1: 

107-12, 130-34. 
Brunt, D. (1939). Physical and Dynamical Meteorology. Cambridge, 

University Press. 
Buchan, Alexander (1869). Interruptions in the regular rise and fall of 

temperature in the course of the year. J. Scott. Mel. Soc. n.s. 2: 4. 
Carruthers, J. (1941). Some interrelationships of meteorology and 

oceanography. Q,. J. Roy. Met. S. 6*7: 207-46; contains many valuable 

references with regard to the Gulf Stream, the effect of wind on sea 

disturbances, etc. 
Dioht, F. H. G. (1931). Thunderstorms of Nov. 2, 1930. Q_. J. Roy. 

Met. S. .57; 101-03. 
Dunbar, \V. (1942). Abnormally high and low daily mean temperatures, 

Kilmarnock, 1902-41. Q..J. Roy. Met. S. 68: 287-92. 
Gold, E. (1947). Weather Forecasts. Q.J. Roy. Met. S. 73: 151-85 
Hawke, E. L. (1937). Buehan's Days. London, Lovat Dickson. 
Lamb, H. H. (1950). Types and spells of weather around the year in the 

British Isles. Q.. J. Roy. Met. S.. 76: 393~43 8 - 
Shaw, Sir Napier (1926-31). Manual of Meteorology, vols. 1-4. Cam- 
bridge University Press. 

chapter 5 


. . Infant winter laughed upon l/ie land 
All cloudlessly and cold: wlien I, desiring 
More in this world than any understand 
Wept o'er the beauty . . . 


It is certainly a fact that so various are the effects of 
weather, light and atmosphere that the landscape 
in England, with its heights and distances, is 
never two days alike. 

L. C. W. Bonacina 

Nothing is more refreshing or reassuring to the English mind than 
the contemplation on a clear day of the pageant of the sky from 
rising ground overlooking a wide stretch of country. The restful 
changeability of the scene thus viewed has the same fascination as that 
of a river. There is a quiet underlying sense of purpose, a deeply- 
felt flowing rhythm beneath the fleeting disorderliness and variety of 
aspect. Faintly-apprehended overtones and harmonics abound to a 
far greater degree than in the brighter and more simplified American 
scene with its harsher outlines, its cruder contrasts, its strongly em- 
phasised seasonal tom-tom, mitigated only where higher latitude and 
wider ocean again begin to modify the behaviour of the air. 

Rhythm can indeed be sought and felt in the roar of the winter 
storms, fitly expressed by Sibelius' incidental music to The Tempest. 
Under the howl of the winter wind the more northern dwellers in our 
island, in particular, recognise that such storms are a necessary and 
regular precursor of summer, and preparations still go forward. Storms 
in the north arc not the capricious ill-tempered irregular devastations 
that beset the resigned peasantry of more southern lands; violent 



though they are, they must regularly be expected at some time in the 
colder months. Nevertheless, protection against them lies within the 

j f M A IvT J J 


S O N D J 

f M A 


1 " 1 1 ' 1 1 1 


i | i • 

i i 


— tn^ond i.Wo!e* / ~ 
— Scollond / 
•--Irfood / t .-■ 


\ \ 

\ \ 



1 ' f 


(i ■ 


i : 

II /fStoKo-J 

1 •' 

a ' 

/ * 



4 • 



s / ■ 
— — ' / I 


r t i . i r 


I i i I 

" 1 


Fig. 29 

Mean temperature 1901-1930, England, Wales, Scotland and 

Ireland (from J. Glasspoolc, Q_. J. Roy. Met. S., 1941) 

capacity of the intelligent individual of cautious outlook. Edinburgh's 
profusion of insurance companies can fairly be associated with die 
demands of the climate on the forethought of the Scottish farmers. 
But discussion of such differences as there are between Scotland 
and England can be left to the reader. After the previous chapters 
we arc in a position to explain why the Englishman who contemplates 
his weather out-of-doors sees and feels what he does. Let us consider 

Plate i i 

Derby: January morning. Going to work in a Midland city. Characteristic 

coppery smoke-haze in the morning inversion layer following a clear night and light 

snow-cover. Temperature 22°. 





the characteristic weather which he is likely to observe, month by 

Winter Weather 

The mid-winter month of January may be taken as typical; we 
can then expect a succession of deep Atlantic depressions to approach 
Britain. A new one appears off our shores almost every other day; 
but frequendy they are occluded by the time they reach us. Polar 
air coming from the westward with a long fetch over the ocean 
gives much low cloud and afternoon temperatures of the order of 46 
inland, falling to 38 at night; sca-suiface temperatures in mid- 
winter range from about 50 off Cornwall to 45 in the Shedands. (Cf. 
Fig. 14, p. 70). 'Moderate to fresh south-west wind inland, strong on 
the coasts', is the most usual outlook; if an occlusion passes, two or 
three hours steady rain is followed by a slight clearance but with little 
change of wind or temperature. With a shorter fetch on the Atlantic, 
however, west winds are colder and more unstable; bright intervals 
of from one to three or four hours' duration on the coasts arc marked 
by a pale blue sky, and flying ragged cumulus over the sea which often 
builds up rather heavily over our hills and mountains. Temperature 
ranges between 35 at dawn to 45 by afternoon, tending to be appre- 
ciably lower in Scotland. Skies often clear a little towards evening 
with a rather pale sunset, broken near the horizon by the jagged tops 
of distant cumulus over the sea. In the evening, ragged low clouds 
still drift across the moon. If the wind dies down sufficiently there 
may be a touch of frost inland before dawn. This is a very prevalent 
winter type of weather; with slight variations depending largely on the 
strength of the wind and the degree of instability in the air moving over 
the sea. It is roughly true that a fast-moving current of maritime- 
polar air is more unstable than one which is crossing the sea slowly. 
Hence with a deep depression centred to the northward the stronger 
westerly winds commonly reach gale force on our exposed coasts; 
lashing showers of rain, not infrequently accompanied by hail, fall 
from die towering cumulus over the white-crested sea, greenish-grey 

Plate 12 

Allestree Lake, Derbyshire: fine anticyclonic winter day, March. Rough ice after 
earlier heavy snowfall. Temperature 34 . 

C.B.S. H 


in the changing light. Temperature rises by day little above 40 ; 
little more than two thousand feet above, therefore, the air is close 
to freezing-point and our cloud-bedecked mountains receive fierce 
driving squalls of sleet and snow. The showery clouds march steadily 


Calm lo^ 

Sol. t- 





Fio. 30 
Wind roses for Durham (with 
acknowledgments to R. F. Bax- 
ter, University of Durham) 

= U-J4 

S 25-Jt 
< J8 

across Britain, but as a general rule the showers are considerably less 
intense and frequent once the air has crossed the mountains. 

If a vigorous depression approaches in which the warm sector is 
still present, the country may be flooded for a few hours with tropical 
air coming from the seas as far distant as Madeira. In this type of 
air the surface temperature exceeds 50 even in Scodand, and tem- 
peratures of 55 occur in S. England. Hurrying low cloud overhead. 



stratus or strato-cumulus is bomc on a soft south-wester; almost every 
year, each winter month from November to February gives at least 
one particularly mild day of this type. Even at night the temperature 
often remains close to 50 . Tropical air of this warmth with a high 
humidity and extensive low cloud needs little further disturbance for 
rain to occur; and generally the arrival of such a warm current gives 
very heavy 'orographic rain' in our western mountain districts. Even 
on the mountain tops the air temperature rises well into the forties, 
and the incessant heavy rain quickly removes the covering snow, 
sometimes up to the level of the highest Scottish summits above 4,000 




jr-i i i t.i 

i'S ' 

I. U I ■■■■.- I . tj . . - .1 I .I..I.1 I 1 ,1 ... I ■_!_. 


Fio. 31 
Annual march of Cambridge temperature and rainfall 

Over the country, however, the cloud-sheet gives litde more than 
oriel spells of drizzle and occasionally for short periods if the cloud is 
thin the sun may be visible. Mild days of this kind with occasional 
sun can be very pleasant given shelter from wind, and most of our 
southern resorts have their gardens in which it is possible to sit out 
on such days, although exposure to the soft, but strongly blowing damp 
wind on the headlands and capes may still produce some feeling of chill 
even when the thermometer is over 50 . The advantage of the resorts 
at which the sea lies towards the southward is that a good deal of 
reflected light comes off" the sea during the mid-day hours. This adds 
to the impression of brightness near the coast even though the sun 
itself may be only intermittently visible through the thinner parts of the 
cloud. We have already noted that maritime-polar air which 
approaches this country from the S.W. after a very long fetch over the 
Atlantic differs little from tropical air as regards its surface tem- 
perature and it will readily be seen that this particularly mild version 



of mP is considerably more 
frequent on the south coasts 
of Ireland and of England, 
than, for example, in north- 
ern Scotland. (Cf. Fig. 15, 
p. 71). 

In some part of January, 
however, wc can generally 
expect that depressions will 
for some days be fended off 
when the high-pressure area 
over the Continent increases 
in intensity ; and briefer spells 
when for a day or two the 
air over England falls quiet 
between two depressions. If 
the continental High in- 
creases, the most usual accom- 
paniment is cP air with 
characteristic dull skies and 
a light to moderate south- 
east wind; we have already 
seen why, towards the margin 
of a winter high, dull over- 
cast weather is wont to occur. 
On such days maximum 
temperatures fall below 40 ; 
minima arc close to freezing- 
point but do not generally 
fall below it unless the sky 
happens to clear; which it 
may do if the stream of air from the Continent becomes very slightly 
drier. More rarely, two or three sunny mild days occur, with light 

Plate 13 

Marlry Common, Surrey: birch trees after an ice-storm. March. Raindrops 
from warmer air al a high level fell through a stratum below trcczing point, and 
froze on impact. The warmer air advancing from the south was unable to displace 
the colder air and while rain fell heavily nearer the Channel, north of the Thames 
there was very heavy snow. 

Fig. 32 

Noon, 18 January 1947. Fine mild winter 
day; note fog at Paris. Very fine, clear and 
quiet interval in the Hebrides within the 
wedge of high pressure; sec also Plate 9, 
opposite p. 78, taken on this day (sec p. 5) 

F. 6'oWriMp 





frost and valley log at night, if an anticyclone moves north-eastward 

from the Azores and is temporarily centred over N. France; the air 

coming round such a high is mild and rather moist and it is always 

touch and go whether low cloud does or does not develop in it. 

(Cf. Fig. 32 and PI. 9). In some years several depressions follow 

tracks across the Midlands, 

up-Channcl, or across 

France, usually when 

pressure is high over the 

Scandinavia-Baltic region. 

In this event cold contin- 

ental air arrives north of 

the track of the centre, and 

precipitation may take the 

form of continuous snow 

for some hours. Under 

such circumstances it is by 

no means unusual to find 

that Southern England 
lies in a warm sector with 
temperatures approaching 
50 , while Eastern Scot- 
land is experiencing a 
heavy snowfall. Such a 
situation is illustrated by 
the adjacent chart. 

Depressions crossing 
Great Britain in this 
manner sometimes move 
along the flank of the continental High towards the Baltic; but if 
the High is well-developed they slowly fill up in die North Sea. 
Squally cold raw north-easterly winds of Continental origin dien pre- 
vail on our east coasts; these are raw on account of moisture and their 
rather low temperature, but after crossing the open North Sea normally 
give readings well above the freezing-point on the coast. Showers in 

Plate 14 

Woodbridce, Suliolk: nearly cloudless May forenoon. Anticyclonic weather with 
stable surface air near the North Sea coast. 

Fig. 33 

Synoptic chart for l8h., 13 March 1947. 
showing thai, with a low crossing the Midlands 
in March, southern England lies in a warm 
sector, whilst eastern Scotland is receiving cold 
continental air, giving snow towards the cast 
coast (For notation, sec p. 5) 



such air may be of cold rain or sleet. From what has been said, how- 
ever, it will be evident that air from the north-east on the margin of 
a low, with a surface temperature of say 37 at Tynemouth, will give 
much cloud and probably sleet showers; but at an altitude of 1,000 
feet, temperature will be about 33 and heavy snow will fall. Much of 
the rapid increase with altitude in the frequency of snowfall on our 
north-eastern uplands can be attributed to this sequence of events. 

Rather rarely, a depression moving from the Atlantic into the 
North Sea brings down a current of really cold maritime-Arctic air, 
in which the strong, unstable north winds give very disturbed skies and 
vigorous snow squalls on all our exposed uplands towards the north- 
cast coast. It seems that later in the winter, about the end of February, 
such outbursts are more likely; February 1955 was notable. 

We have thus built up our average January. At a Midland station 
such as Keele or Leicester, for much of the month air ranges between the 
milder and the colder varieties of maritime polar. It is rather windy 
with intermittent clearances especially towards nightfall, and tem- 
peratures fall in the range 35°-45°. On one or two mild misty days 
50 is just exceeded; a few days of dull cold weather give values in the 
thirties. On an exceptionally clear night in the 'wedge' following the 
passage of a low a minimum near 20 is recorded inland ; there is per- 
haps one day (also in a wedge) which can be called sunny almost 
throughout, with about six hours recorded duration. (Cf. Fig. 32, p. 98 for 
18 January 1947, die wedge being over Scotland). Perhaps on ten 
mornings the minimum has fallen to 32 or below. Typically we 
should record by the end of the month that snow fell in small amounts 
on about four days, and on a few days during which the continental 
air prevailed, we should note it as lying on three mornings. Rain 
would have fallen on about sixteen days; mosdy in rather small 
amounts but with two or three bigger falls of the order of half an inch 
when a rainy sector of a vigorous depression moved eastward across 
the country. Bright sun might have been recorded inland for forty 
hours of which three days might between them give fifteen; the rest 
would most frequently be recorded as one or two hours on a number 
of days, chiefly those with westerly wind blowing and clearances of 
some length between showers. 

In the diagram below, based on data from Cambridge, we may 
observe the contrast between die trend of temperature in an excep- 
tionally mild January (mean temperature 44-5° in 1916) and a very 



cold January 1940. In 1916 aldiough the prevailing air supply was 
exceptionally mild die mean daily range differed litdc from that 






— r— 1 — r— 1 I 1 — 1 I I i I — I — I • I t I — I — I — I — I — I — I — I — 1 — I — r— I — I — 1 — r— 

5 JO 15 20 25 30 o 

COLD January 1940 : Mean, temperature 29-3 





r— r— 1— t— 


5 10 15 20 25 30 „ 

WARM January 1916: Mean temperature 44 5 

Fio. 34 

Comparison oflhe daily maximum and minimum temperatures for January 1916 

and. January 1940 at Cambridge 

which can be expected in a normal January when maritime-polar 
air predominates with a good deal of cloud. The cliicf characteristic 
of milder Januaries is that continental air scarcely reaches this country 



at all, and on several occasions tropical air with maxima over 50 may 
be experienced; hence in an exceptionally mild mid-winter month, 
the mean may reach 44 even in die eastern coundes, with an average 
daily maximum approaching 50 . In 1916 at Cambridge the mean 
daily maximum and minimum were 50-8°, 38-2°; normal being 

44-7°, 33 -9°- 

Below the main diagram, however, a second diagram shows the 

trend of temperature in the really severe January of 1940; the above 
means at Cambridge were 293°; and 35-8°, 22-8°. Over the greater 
part of the Midlands and even on the south coast as far west as the 
Isle of Wight, the mean temperature for the month was below the 
freezing-point and in places where snow fell early in the month it 
covered the ground almost throughout. Pressure was generally high 
and for a great part of the month the highest pressures lay to the north- 
ward; the country was flooded with dry conrincntal air in which on 
some days even the maxima remained below the freezing-point despite 
considerable sunshine. In the middle of the month the passage of a 
depression eastward across France gave the coastal belt of South 
Eastern England a classical snowstorm (sixteen inches at Eastbourne) ; 
and on the following nights of clear skies temperatures well below 
zero were recorded inland in Kent and Sussex where the snow cover 
was deep. But on the same night in East Anglia dicre was scarcely a 
powdering of snow. The resultant distribudon of minimum tempera- 
tures for the clear calm morning of January 20th, 1940 is shown on 
p. 169 (Fig. 54) as far as the figures allow. It gives an instructive 
example of die effects of urban and coastal location, of sandy soil in 
the Breckland and of position on rising ground in the Chilterns. 

The frequency with which the country is flooded with moist 
surface air is the cause of the high average cloudiness; if we put all 
the observaUons together we shall find that most stadons average 
between seven and eight tenths of the sky covered with cloud at the 
observing hours. For the months of November, December and 
February conditions are broadly similar. But as the average sea 
temperatures fall off Cornwall from nearly 55 in November to 48° 
in February, and correspondingly elsewhere (50°-44° at Wick, 50 to 
below 42 ° off Norfolk) the average temperatures prevailing during the 
onset of Atlandc air, whether tropical or polar by origin, steadily 
decrease. Hence in west-wind weather in February there is a greater 
chance of night frost occurring than in November. Tropical air with 



almost overcast skies in early November occasionally comes inland to 
give maxima over 6o°; the exceptionally warm days and nights from 
November 20th-23rd, 1947 have already been mentioned (pp. 72, 73). 
Polar air coming by a shorter sea route in February often gives day- 
time temperatures below 40 and sleet may fall through it even on the 
coasts, especially in Scotland, widi sharp snowfalls on high ground. 
In November, on account of the warmer sea surface, this is less likely 
to happen. Hence, at lower levels south of the Highlands the frequency 
of snow in Februaiy is about four dmes that in November. 

To offset the effects of the cooler sea, however, by mid-February 
the sun is more powerful than at any time in November and in quiet 
clear weather die daily range begins to increase considerably. More- 
over the air comprising the warm sectors of depressions which reaches 
our shores in November has a higher vapour content by reason of its 
higher temperature; November and December, therefore, can be 
expected to give heavier rains than January and February in a normal 
year. On the whole, December is the month in which the deep 
Atlantic lows most frequendy affect our western coasts; in January and 
February when the Continental High is most likely to develop and 
spread, some of them may be fended off. Hence while great gales may 
occur in any winter month, the figures taken over many years show a 
slight maximum in December; and we can justify our view of dark 
December as the stormiest month. Such deep depressions with their 
floods of warm moist air surging from die wild Atlandc over our 
western mountains give immense falls of rain in certain well-known 
localities. There, convergence of the air streams in constricted mountain 
valleys enforces additional ascent of air to add to that resulting from 
the mountain barrier itself. (Fig. 41, p. 131). 

Late February and onward into March is the season at which the 
sea is coolest, while in the Arctic February is often a cooler month 
than January. It is then that maritime-Arctic air sometimes descends 
on us with exceptional cold; this is the more marked in Scotland, as 
one might expect nearer to the source. Towards sunset the northern sky 
is often greenish by contrast with the remaining fragments of cloud; 
as the air falls quiet under the stars, with the aurora flickering above 
the northern horizon hard frost follows and the Highland valleys often 
experience their lowest minima of the year ( — 13° at Braemar, 1955). 

Such outbreaks of cold air behind depressions following southerly 
tracks may also give a good deal of snow, not oidy in those eastern 



uplands where we normally expect it for orographic reasons, but also 
in unexpectedly mild locations towards the west. 

Rather rarely, particularly heavy falls of snow are experienced for 

example in Western Ireland. These are attributable to exceptional 

developments in a stream of Arctic air. Normally as we have seen 

instability showers develop widely when such a cold stream debouches 

over a warmer sea. Sometimes the result of a cluster of such showers 

is that pressure is lowered sufficiently over an area of sea for the air 

to begin to follow the isobars round it; on the weather map we find a 

sinister little area of low pressure, perhaps in the region of the Faroe 

Islands. Like an eddy it moves southward in the general stream of air; 

within it, as in all low-pressure areas, there is considerable convergence 

and ascent of moist air, with consequent precipitation over a wide area. 

But in such a system there is nothing but cold air, and the result in 

general is an unexpectedly lively snowstorm. The blizzard of April ist, 

19 1 7, in Western and Central Ireland; the heavy snowfalls of 16 May 

1935 and 10 May 1943 in Lancashire; the fall of eighteen inches of 

snow in Suffolk on 27 April 1919 may be mentioned. Such depressions 

moving southward gradually draw in different air supplies and acquire 

more defined fronts; an example is shown for 10 May 1943 when the 

Northern English uplands received their heaviest snowfall of the 

'winter'. (Fig. 36, p. 113.) Repeated instability showers with a strong 

north wind of Arctic origin gave very heavy drifting snowfalls in 

Northern Scotland in February. In sharp contrast a rare day of dry 

cloudless south wind gave 65 in London on Valentine's Day in 1961. 


Chapter 5 

Bilham, E. G. (1938). The Climate of the British Isles. London, Macmillan. 
Bonacina, L. C. W. (1937). Q.. J. Roy. Met. S.,63, 483-90 Constable as 

a painter of weather. 

('939)- 'bid. 6j, 485-96. Landscape 


(1940). ibid. 

weather and climate. 

(1941). ibid. 67, 305-12. The scenic 

approach to meteorology. 
Hawke, E. L. (1937). Buchan's Days. London, Lovai Dickson. 

66, 379-88. Scenery, 



Bursts from a rending East inflows 
the young green leaflets' harrier 
Meredith: Hard Weather 

There was such deep contentment in the air 
that every naked ash, and tardy tree 
Tel leafless, showed as if the countenance 
with which it looked on this beautiful day 
were native to the summer. 


Long ago the remark was made in a March issue of Punch that 
"spring has set in with its accustomed severity". This expression 
indeed was used in a letter by Coleridge written at the beginning of 
May in 1826; but he in turn may have known that Madame de 
Sevignc used a similar phrase in 1689. Adopting the division of the 
year into four seasons of equal length, a division likely to appeal to 
most meteorologists in mid-temperate latitudes, it is undoubtedly best 
to consider March, April and May as comprising the spring. But 
surrounded as we are by the sea the rise of temperature with the 
lengthening day is slow and frequently suffers many setbacks; hence 
in Britain we have a very long season during which one after another 
ol" the familiar harbingers of spring appear. Many of our flowers 
derive from wild ancestors with an open deciduous woodland habitat, 
where they quickly responded to the increased light while the trees 
were still bare. Hence such flowers as crocus and daffodil, anemone 
and bluebell successively appear to remind us of the approaching 
warmer season. While in some years a venturesome crocus may appear 
a month early in mid-January the daffodil is equally capable of appear- 
ing a month late, in mid-April. March is indeed a particularly variable 



month, although not from the point of view of mean temperature; 
in that respect its possibilities are surpassed by December, January 
and February. The range of possibilities in March (and in Scotland, 
April) is however probably more effective for an easily-understood 
reason. Our prevailing vegetation characteristic of the temperate 
lands of Europe and parts, at least, of North America is such that 
growth begins and is maintained as soon as the mean temperature rises 
above a figure between 42 ° and 43°F. Climatological studies made in 
France gave this figure as 6°C. (42-8°F.). Arising out of such studies 
and that of others the Meteorological Office in Britain has for many 
years worked out and published values of "accumulated temperature" 
above 42 for each week, month and season. These figures represent 
the extent to which in any week or series of weeks the temperature 
has been in excess or deficit so far as plant growth is concerned, with 
a view to correlating the extent of the progress made by crops with 
the weather conditions. 

It does not matter for the purpose of the present argument which 
figure we adopt around 42 °; it is more important to realise that the 
mean sea-level temperature of March in England is 42-9° (period 
1906-35) and of April in Scotland 44-2° for the same period. Hence 
it is immediately evident that small fluctuations of the mean tem- 
perature of these months will have a big effect as regards the progress 
made by vegetation. If March in England is severe, with a mean tem- 
perature of 38 , it is reasonable to assume that the average daily 
maximum will be about 44 , and the night minimum about 32 °. Thus 
taken over the month as a whole the temperature will on the average 
only exceed 42 for about two hours out of the twenty-four. But in 
a mild March with a mean temperature of 46 and an average daily 
range from 52 ° to 40 , by far the greater part of every day and night 
will be above 42° when averaged over the month and the proportion 
of hours during which growth is likely to proceed without check may 
well be ten times that of the cold month. 

Note too that we are here discussing the temperature of the air, 
taken by standard methods. Patches of ground beneath the shelter 
of a tree, or adjacent to a warm wall, do not radiate so freely at night, 

Plate 15 

Shropshire Plain and Wrekin from Wcnlock Edge: April. Cloudless warm anti- 
cyclonic April weather. Clear air free from smoke-pollution. 

plate 15 




Cyril j\'ftrbrrry 

Cyril .V.n !>,rrj> 

and the minimum air temperature to which plants are subjected at 
night may be appreciably above that in a less protected situation. 
Thus we can find, especially in a cold month, marked variations even 
in the average garden in the response made by growing plants. If we 
extend our survey to cover the sheltered south-facing hillside nooks, 
the marshy frost-hollows with their coarse grass, the shady woodlands 
and the windswept ridges of this varied country of ours we shall find 
that the appearance of the first signs of spring and the progress of the 
season shown by vegetation depend to a remarkable extent on these 
micro-climatic differences, as we may call them. These are not as 
large in magnitude as in some countries elsewhere; but in the earlier 
spring months of March and April the differences in maximum and 
minimum temperatures from place to place arising from exposure, 
relief, soil and shelter arc the more effective because they range on 
cither side ol several critical mean values (see also Chapters 8-10). 

Spring sunshine is quite powerful and effective if the air is quiet. 
The sea, on the other hand, is at its coldest. To this we must add the 
fact that in a normal year there is still much ice in the Northern 
Baltic and that in the Greenland and Labrador seas the ice is generally 
more widespread and closely packed than at other seasons. When it is 
melting, a layer of cold and relatively fresh water from the ice lies 
on top of the salt water beneath by virtue of its lower density. We have 
already seen that in winter months the Siberian anticyclone dominates 
the surface circulation of the air over Asia and Eastern Europe. 
Towards spring it becomes less intense; but at the same time the 
extensive snow-covered Scandinavian Highlands and the icy Baltic 
remain as a relatively cold area by comparison with Europe to the 
southward. We find a well-marked tendency for anticyclones to 
develop and persist over the Baltic lands. We might justly attribute 
this in part to the fact that the movement of depressions depends on 
the existence of well-marked fronts. In the spring months air of 
Atlantic origin making its way into West and Central Europe, by this 
time generally quite free from snow, is often distinctly warmer than 

Plate 16 

a. SWAP Fell. Westmorland: sunshine and shower. May. Cool westerly weather 
with showers developing at intervals over die Lake District in rather unstable 

maritime-polar air. 

0. Shrewsbury, Shropshire: warm anticyclonic April. Evening, by the Severn. 



that which has been cooled by radiation over snow or ice farther north. 
Fronts therefore are often to be found somewhere across West Central 
Europe; and our islands often lie on their northward side. 

Pressure is also (at sea level) generally a little higher over Greenland 
throughout the winter and spring, than it is over the open Norwegian 
sea or the North Atlantic. In spring this Greenland High frequendy 
spreads across the sea to eastward. In such a case cold continental- 
polar, and even Arctic air reaches us pcrsistendy, in the form of the 
very common and familiar "east winds of spring"— it would be fairer 
to recall that their actual direction varies between N. and E. They 
share however the characteristic features of the continental air of 
winter already described in that they are unpleasantly cold and 
penetrating; as the proverb says "neither good for man nor beast". 

Scandinavian and Arctic highs however are not persistent through- 
out and do not always become established at the same date. In many 
years the Atlantic depressions continue to develop on the polar front 
south of Iceland and to move with little or no interruption on the 
normal track towards the Norwegian coast. In this event cool westerly 
maritime polar winds again predominate with if anything slightly 
more stable conditions (owing to the cooler sea off our shores) than in 
mid-winter. A fresh blue sky with patches of small cumulus or cumulo- 
stratus, and a cool westerly breeze from the sea will be recalled by 
many as characteristic of March; the trees are still bare, but the sun is 
bright and distinctly warm out of the wind. The grass begins to grow, 
larks sing, the cattle are turned out for a period each day; daffodils 
flower here and there, crocus is abundant, the willows are turning 
green. Daytime maxima reach 50 or just above. At night the 
thermometer may fall to below 35 , but in many places, especially 
where the air is still moving, there will certainly be no sign of ice. 
The mean temperature of course is still close to that of the sea surface 
from which the air is coming; it lies in the neighbourhood of 43°-46°. 

In such circumstances there is in March a decidedly significant 
difference between Cornwall, where the sea temperature by now is 
down to 48°, and the Outer Hebrides where it is 44 . The shorter 
distance travelled by the maritime-polar air to the Hebrides and the 
cooler sea combine to render the normal west-wind day as described 
above about five degrees cooler in the north. This means that at 
2,000 feet on the Brecon Beacons where the probable afternoon 
temperature is about 39 , sleet will not be observed in a passing 



shower. But at 2,000 feet on the N.W. Highlands the day-time average 
will be about 33 ; and frequently the ground at this level will be white 
with snow whenever the normal cool westerly weather prevails. This 
is reflected in the figures; the mean frequency of occurrence of snow- 
cover in March at 2,000 feet is about five times as great in Sutiicr- 
landshire as it is in South Wales. 

Tropical air if it reaches us comes from the sea at its coolest, and 
hence tends to carry a good deal less water-vapour than it will for 
example in October. As tropical air masses, either directly or after 
occlusion, often give us our longer-duration rains we tend to get less 
rain in March, and still less rain in April than we expect in January 
or February; a further cause of this decreased rainfall lies in the 
slightly greater stability of the 'showery' west wind above-mentioned, 
and in the frequency with which anticyclones develop to the north- 
ward, as we saw. All these factors together make the average April 
one of the driest months of the year; although the frequency of measur- 
able rain throughout the spring months does not decrease at all so 
markedly as the quantity, by comparison with autumn and winter. 

March is then a variable month in which as a rule we can generally 
expect several days of dry, cold north-easter, with some strato-cumulus 
cloud as before near the east coast, especially towards the southern 
North Sea where a north-easter has a longer 'fetch'. There will be a 
number of days of cool west-wind weather; if a cloudy warm sector 
crosses us, the air temperature (52°-53° by afternoon) is no higher 
than in December or January. But if the sky happens to clear for a 
time, even with the wind blowing freshly from the S.W. the sun may 
give us an afternoon maximum inland of 56°-57° (Midlands) on one 
of those mild March days, and at night temperatures will scarcely fall 
below 45 . The dry cold days give (with continental air) maxima of 
the order of 45 inland, and minima if the sky clears of about 30 . 
The west-wind days we have already mentioned. 

Minor fronts in the prevailing maritime air-stream give much of 
the rain; but in this respect March varies greatly. In a year with a 
vigorous Scandinavian high and disturbances crossing the country well 
to the southward along a well-developed front between the cold con- 
tinental air and that of the Atlantic, there may be a good deal of snow, 
especially in the north-east. Moreover the existence of high pressure 
in the north means that in almost every March, and quite often in 
April, there is one real outburst of maritime-Arctic air, giving the 



snowy 'instability showers' we have already described on all the 
exposed coasts and uplands, sometimes as far south as the E. Kent 
hills. By reason of the less distance the air has had to travel over the 
warmer sea, such showers of snow are very normal in E. Scotland 
where they have acquired the name of 'the lambing storms'. The 
frequency of snow-falling and of snow-lying varies more noticeably 
from south to north in March than in any other month. 

Generally in the clear night skies and calm air following such a 
bout of north wind severe frost is recorded at least once in the month. 

Fig. 35 

.Comparison of the mean pressure at 7 a.m. for March 1937 and March 1938 

(in millibars) 

In the Midlands the average extreme minimum for March is about 
23°, only a degree or two warmer than that for January or February. 
The most notable departures occur in March when anticycionic 
weather is really persistent. We may illustrate very well by reference 
to the very cold March of 1937 and the very warm one of 1938 in the 
above charts. Given a warmish air supply off the margin of a High 
over France and dry ground extraordinarily high day-time maxima 

Plate 17 

Princes Street, Edinburgh: early May. Springtime: almost calm, slight haze, 
very light air from S.E. Cautious retention of coats by older Scotsmen. 

PL ATE 17 


Cyril Xtubtrry 


f .'./.-/ \'acbtny 

Erie Hoiking 



above 70 sometimes result, although the ground beneath the surface is 
still relatively cold and little warmth is conducted to the surface on 
clear radiation nights, so that night frost still occurs. 74° in Northum- 
berland in 1957, 73 at London in 1961, are notable. 

With regard to April air masses practically the same remarks may 
be made as for March; but by this time, whatever happens at night, 
the sun is strong enough by day to send the maximum temperatures 
well up the scale into the fifties in the South of England. Towards the 
coast the effect of the cool coastal sea breeze on afternoon tempera- 
tures begins to be noticeable on quiet days, especially adjacent to the 
North Sea. Moreover we find that at this time of year air crossing the 
cool sea either from the warmer land of the continent to the S. and 
S.E., or from the warmer ocean away to the S.W., tends to become 
decidedly stable in its lower layers; it may give extensive cloud, but 
there is little chance of precipitation, in contrasi to the conditions 
of November or December. As we have seen the Scandinavian High 
quite frequendy develops, and an early April north-easter across the 
cool North Sea is very little warmer than that of March; but as the 
month wears on the chance of such chilly surface air derived from 
Scandinavia steadily decreases, at least in a normal year. 

Frost still occurs at night in April given clear skies and a cool 
air supply; and if the air is of maritime-Arctic origin, coming down 
behind a depression in the North Sea, the resultant night frost may 
still be quite severe. At habitable levels the chance of a day-time 
maximum below 32 is now very remote even in the North. Snow 
can still be expected to fall, at inland stations in Southern England, 
on one day in a normal April usually in the form of a shower; it is 
distinctly unusual in the south for snow to cover the ground. In 
Scotland, an April day with snow-cover must regularly be expected 
over the greater part of the low country; and at high levels the fre- 
quency is still considerable. Notice again how the south-north gradient 

Plate 18 

a. The Clyde Estuary from above Greenock, Renfrewshire: late afternoon. May. 
Warm dry south-easterly wind, slight build-up of cumulus over distant hills: 

depression to westward of Ireland. 

b. Gorple Reservoir, Yorkshire West Riding: early summer on the Rritstone 
Pennmcs near Burnley, June. Fine weather cumulus with light N.W. wind (hence 
excellent visibility); cotton gTass seeding; characteristic June anticyclone well 

to northward in the Atlantic 
C.B.S. I 





of mean temperature is greatest in March and April. The most rapid 
change in the (eel of the season can be got at that time, a fact which 
the Edinburgh spring visitors to the sheltered South Devon coast 
regularly perceive and enjoy; the ultimate cause lies in the appreciable 
difference between the sea temperatures and the fact that dicy lie 
close to the critical value for plant growth. 

April then like March is a mixture of dry cool weather with a good 
deal of sunshine, somewhat showery cool westerly weather with long 
fair intervals, and now and then the more extensive cloud and rain of 
a transient depression. The soft moist growing weather beloved of 
the farmer is fairly frequent, associated as a rule with a slow-moving 
current of humid air round the margin of a High over France or 
Germany. Such air is frequendy 'returning maritime-polar'; it may 
be moderately stable. A typical soft April day of this type gives 
afternoon temperatures of 58 , and probably clearer skies and more 
sun dian the broadly analogous type of day in March. 

Dry clear warmth in an April High moving over the country may 
give us a very large day-night range of temperature; in Southern 
England, about every other April gives one day with a maximum just 
over 70 . In Scotland the narrower extent of land i.e. closer proximity 
to the sea rarely permits such high temperatures even in quiet sunny 
anticyclonic weather. To get really warm days in Scotland in April 
it is usually necessary to postulate a High over the region of Germany 
with a drift of dry and fairly warm air of Continental origin all the 
way up to Scotland over land; even then, April maxima over 70 
occur rather rarely. By contrast the fearful effects on temperatures 
of the deep snow-cover which covered the north early in April 191 7 
may be mentioned. The night after the snowfall was clear and calm; 
radiation was rapid, and at Penrith the screen minimum fell to 5 . 
Many places in Scotland and North England recorded below io° and 
there were some unofficial reports of minima below zero. As die 
majority of Aprils since 1930 have been warm we tend to forget the 

In May it is still possible for night temperatures to fall danger- 
ously low, given clear nights. This is one of the most fundamental 
features of our climate, of the greatest importance to fruit-growers, 
and also to all who arc interested in the production of early vegetables. 

The general conditions arc much the same as in April; in some 
years depressions continue to follow the normal Atlantic routes, in 

others they are temporarily checked by the development of a large 
High, quite frequently over the Icelandic-Norwegian seas. Sometimes 
however the Azores High begins to spread northward, and fine clear 
dry west-wind weather is established for a period in S. England, with 
stable air and little chance of rain. The High may gradually spread 
right over Britain. 

If an anticyclone in which the air as a whole has been derived from 
warmer regions is centred over these islands with clear skies, the May 
sun gives high maxima; 8o° is occasionally reached in S. England before 
the 10th of the month. In the north and in Scodand however 8o° is 
very rarely touched before the end of May. Further the ground 
is often dry after a fine April, and hence radiation on clear nights is 
effective. But anticyclones 
may equally well spread over 
us from fardicr north. In the 
event of a cool dry northerly 
or north-easterly current 
spreading over us associated 
with an advancing high 
behind a depression which 
has moved eastward, day- 
time maxima are governed 
by the fact that the mean 
surface temperature for the 
whole day can differ little 
from that of the North Sea 
over which the air has come. 
Hence while the centre of a 
High still lies to the north- 
ward, the cool north-cast 
wind in S.E. England is likely 
to give a maxima little above 
55 , and minima fall well 
below 40 at night over most 
of the country. With minima 
at representative country sta- 
tions of the order of 35 c -38° 
many localities in valleys and 
hollows, especially with 

Fio. 36 
Snowfall in May; 0700 hrs., io May 1943 

u 4 


lighter soils, may experience frost. In a later chapter we shall sec how 
important this sporadic incidence of May frosts can be in a normal year. 

Arctic air is still capable of reaching us and giving snow showers 
down to sea level in ScoUand and N. England; as we saw in the pre- 
vious chapter. But with the more powerful sun and the consequent 
warming of the land, such a cold air stream generally warms up suf- 
ficiently in the surface layers to make May snowfalls distincdy infre- 
quent in the South, and snow-cover even for a brief period is very rare. 

On the other hand we can recognise the advent of a new and very 
characteristic type of weather on and near the coasts, especially that 
of the North Sea. If pressure is fairly high over die region of Germany 
and low towards Iceland or the Bay of Biscay the air stream reaching 
us from east or south-east comes from the relatively warm continent 
on to the cooler North Sea. Sometimes, coming from the south across 
France, it crosses the cooler Channel. If such a current leaves the 
Continent rather humid in its surface layers, as it may well do after 
a spell of showery weather and extensive damp ground, it is cooled 
over the sea and all the requisites are present for 'advection fog'. In 
the case of the North Sea it is generally found that in spring and early 
summer the coolest waters lie between the H umber and Aberdeenshire 
along the English and Scottish coasts; on the Dutch-German coasts 
the surface water is three or four degrees warmer. Hence we find that 
air, already rather humid, is flowing over a cooler sea and so becoming 
relatively more moist soon after leaving the Dutch coast; and fog or 
very low cloud begins to develop, becoming quite thick on our coasts 
where the saturated air is crossing a belt of still cooler surface water. 

Plate Va: Clouds over the Dorset coast, afternoon, mid-March, 1925. Cumulo- 
nimbus and a passing thundery shower inland to eastward ; unsealed with unstable 
polar air and excellent visibility. 
b: The Isle ol Man Irom Galloway, September 1937. Excellent visibility 
in the quiet air with lilUe ground heating underlying extensive high cloud. Air 
flowing smoothly over the island building up strato-cumulus with base about 1000 feel 
above the Manx summits (Snaefell, 2034 ft.) but unable to ascend to greater heights. 
Slight mirage of the low hilJs at the N. end of the islands over the relatively cool *ea. 
" In his tone course the shepherd oft will cause 
And strive to fathom the mysterious laws 
By which the clouds, arrayed in light or gloom 
On Mona settle, and the th/ipes assumr 
Of all her peaks and ridges." 



The air-mass leaving the Continent is thus cooled to a depth which 
is often of the order of 1,000-1,500 feet. By the time the air reaches 
our coasts it is virtually saturated, with consequent fog formation 
throughout this depth. Here then are our classic advection fogs; 
but we must note that with a slight increase in wind speed and con- 
sequently increased surface turbulence, the base of the fog is found 
slightly above the surface, perhaps from 200 to 500 feet. The process 
has already been described in Chapter 4. 

The result is the formless thin grey-morning fog, or very low cloud, 
sufficiently conspicuous in Edinburgh and E. Scotland in the spring 
and early summer to acquire its local name of 'haar'. 'Sea-fret' is 
commonly heard farther south; in Lincolnshire and Yorkshire 'sea- 
roke', 'roke' or 'north rokc' arc heard. (Icelandic reykur, Danish r»k= 
reek, smoke or fog). In some years with a persistent High there may 
be several days when it lasts throughout, e.g. May 1935. In general 
however it shows some sign of lifting and disappearing during the day- 
time, especially if there is a clear sky above. For, as the sun rises during 
the forenoon, radiation penetrates the low cloud and reaches the 
ground from which the surface air over the land is warmed; moisture 
is absorbed, and after a time the low cloud disappears. Two other 
factors aid the process. As the land warms, the sea breeze increases in 
speed ; this may give more turbulence over the land surface, mixing the 
saturated air with that which is unsaturated above. Off the coast it 
also leads to some slight descent of the dry air above, which may be 
sufficient to absorb some of the moisture in the surface layer. More- 
over, as soon as direct sunshine begins to penetrate to the ground and 
can thus warm the ground well, the fret generally disappears quickly. 
If on the other hand there is a slowly spreading cloud-sheet at a higher 
level, obscuring the sun, the surface sea-fret remains very obstinately and 
the gloomy damp grcyness is indeed depressing. In general such a spread- 
ing cloud-sheet, at a higher level, is the precursor of a front ; eventually 
the sea-fog beneath it will slowly be removed after the onset of rain, 
followed by the arrival of a different type of air. Interesting experiments 
can be carried out by any amateur with two or three thermometers 
during the dissipation of fog with a clear sky above (cf. Fig. 3, p. 27). 

The rather sudden appearance and disappearance of this extremely 
characteristic and dismally chilly veil of low cloud has for long been 
observed and in many parts of the coast is proverbially associated with 
the tide. In 1 750, Dr. Thomas Short of Sheffield commented on the 



'tide-wcather' of Lincolnshire. "This sort of weather will change with 
the tide" is commonly heard; in general however such a statement is 
not justified, as anyone who notes the times of high and low water and 
correlates them with the variations of the sea-fret will soon observe. 
Locally however, for example, in the Essex estuaries where the ebb 
reveals a wide stretch of sand, the drying of the sand and the ease with 
which its surface warms by radiation from above probably does play 
some part with regard to the moisture of the adjacent surface air, and 
hence the local variations in the thickness of the cloud. The fore- 
casting of the spread of night radiation fogs in certain sandy estuaries 
is affected by the extent to which the tide covers die sand; and it is 
not unreasonable to suppose a similar effect with regard to the advection 
fog so characteristic of the late spring and early summer. Patches of 
cold water may also play a part, as is well known by airmen near the 
Solent. Such associations with the tide however can at best be very 
local; and it would be wrong to force such conclusions on the rocky 
North Yorkshire coast, for example, in a region where the dde recedes 
but a short way from the cliffs. 

Coastal fogs of a similar type are by no means unusual in the region 
of the mouth of the Channel, notably in June and sometimes in July 
or even later. A light warm southerly wind coming round the flank 
of a large continental high from across W. France is frequently accom- 
panied by widespread fog or very low cloud along the South Devon 
and South Cornwall coasts. It may spread up the west coast to such 
peninsular regions as that of the Lleyn in N. Wales. More commonly 
however as we go northward the wind increases and the result is the 
extensive low stratus at 1,000-1,500 feet accompanying a mild but 
curiously relaxing south to south-west wind in the Lake District, for 
example, or in S.W. Scodand. Anyone who has tried to walk on such 
a day in Eskdale in early June will appreciate the way in which a 
singular lassitude discourages one from climbing into the formless 
stratus above, although from this type of cloud in the stable surface 
air nothing more than a slight drizzle occurs even in the mountains. 
No doubt if our mountains were high enough it would be possible to 
climb above the cloud into clear dry air. 

The sense of lassitude is probably associated with the fact that 
inland in the valleys the surface air movement is slight; on the coast 
wind speed may be of the order of 8-12 m.p.h. Secondly, the surface 
air is humid and evaporadon from the body is slow, especially with 



little air movement. By June the sea surface temperature on our west 
coasts is well into the upper fifties, the air temperature in the day 
probably lies somewhat above 6o° and there is a curiously oppressive 
sense of warmth from overhead due to radiation through the cloud, 
not merely from the invisible sun but from the overlying warmer mass 
of air. In a later chapter it will be seen that such temperatures with 
high humidity and little air movement are much more discouraging to 
exertion than if the air were saturated and io° cooler. The frequency 
with which such conditions occur in the summer months towards our 
west coasts goes far to explain the relaxing quality of the climate, 
especially in those inlets and deep valleys among mountains in which 
the movement of air is restricted. 

All these early summer effects as we have seen arise primarily 
from the flow of warm air over cooler seas. More, however, should 
be said of June. By this time the Arctic ice off* Greenland has largely 
dispersed; the sun is at its most powerful and throughout the month 
is overhead at noon in latitude 20 or more north of the equator. 
Associated with this northward displacement of the region in which 
incoming radiation exceeds that which is out-going we find that the 
great mid-Atlantic or Azores High tends to take up its most northerly 
station, while at the same time the warming of the land produces, on 
the average, lower pressures to the eastward. Britain in June has a 
very fair chance of lying within the influence of an Azores high, 
or within that of the extensive masses of quiet air called anticyclones 
which appear to detach themselves from it and to move slowly 
north-eastward over the north-west coasts of Europe. In such massive 
anticyclones the air above is warm, dry and clear, pardy on account 
of the subsidence which is necessarily associated with such highs. 
Surface air generally moves from westerly points, from a sea which 
is still rather cooler than the land. Hence the air as it flows over 
the land is warmed, and the chance of extensive cloud formation 
within it is decreased. Inland the strong radiation encourages con- 
vection, and there is still in general enough moisture rising from the 
surface to provide small cumulus clouds in the day-time. But, on 
account of the subsidence of warmer and dryer air at higher levels 
the vertical growth of cumulus is generally checked at no great height. 
We thus get the very characteristic fine-weather cumulus associated as 
regards place with inland locations, increasing a little towards after- 
noon and dying down at night, perhaps disappearing entirely, perhaps 



remaining as fragmentary stratus. (Cf. the photograph on PI. 18b, 
p. in). The sky above is a clear blue; many hours of bright sunshine 
are recorded, and indeed an anticyclonic spell in June can scarcely 
be matched for sheer enjoyment in any other climate in the world. 

On the long run of years the first few days of June appear to be 
one of the most favourite 'spells', especially in north-western districts; 
the great regard in the north for the Whitsuntide holiday owes much 
to this tendency. 

Visibility is generally good wherever the sea is not far away. The 
clear unpolluted smoothly-flowing oceanic air from west or north- 
west gives sharp outlines. In Scotland, where the oceanic air is par- 
ticularly pure, the brilliance of the colouring in fine June weather with 
a High to the westward is unforgettable. Farther inland however, 
where the air has had a longer land travel, the rising currents from the 
surface produce considerable haze, part of which can be attributed to 
dust from the surface and part to the smoke of our industrial areas. 
For example, in a summer anticyclonic spell a light westerly wind at 
Cambridge is often associated with brilliant skies above, but visibility 
at the surface of less than three or four miles in the afternoon. This 
can no doubt be partly attributed to die gradual shift of the morning's 
smoke from the Midland cities, associated with the surface turbulence 
and rising currents to a height of between 3 to 4,000 feet as a general 

The frequency with which such anticyclones develop, the de- 
creased activity and speed of movement of Icelandic and other 
depressions, and the prevailing stability near the coast of surface 
currents moving from a cool sea on to a warmer land, all combine with 
the length of day to make June the most sunny month of the year. 
Indeed even when we allow for the length of day and take out the 
proportions of bright sunshine for each month, June almost everywhere 
shows the highest average percentage of possible bright sunshine; 
though in some parts of the South it is slightly surpassed by May. 


Chapter 6 

Manley, G. (1935). Some notes on the climate of N.E. England. 

Q..J. Roy. Met. S. 61: 405-10. 
Short, Thomas (1750). New Observations on the Bills 0/ Mortality. London. 

chapter 7 


Less fair is summer riding high 
in fierce solstitial power 
Less fair than when a lenient sky 
brings on her parting hour 

From all that has been said it will be clear that no sharp line can 
be drawn between the seasons. On rather rare occasions a per- 
sistent anticyclone to the north of us may prevail into June; cool 
north-easterly winds from Scandinavia cross the chilly North Sea and 
continue to give extensive day-time strato-cumulus cloud towards the 
east of the country. At night the decreased turbulence and clear skies, 
following a day with a maximum temperature in the fifties, allow of 
rapid cooling; and towards the north, inland valley-bottoms can 
generally expect a night or two with minima close to the freezing-point 
in the earlier part of the month. From mid-June to late August 
however, it is pretty unlikely that even the worst-favoured inland 
location will experience a screen minimum as low as 32 ; though they 
have occurred as the table on p. 257 shows, e.g. at West Linton. West 
Linton lies in a broad upland basin 10 m. S.W. of Edinburgh, sur- 
rounded by bare grassy uplands. With the break-up of much of the 
Arctic ice and the warming of the Arctic land-masses, it is scarcely 
possible from June to early October to make any distinction of mari- 
time-arctic from maritime-polar air. 

Throughout the summer months it is still true that we arc more 
often than not under the sway of air masses which have been for some 
time over the Atlantic. Westerly winds prevail; depressions continue 
to move along the polar front, that is from Iceland to N. Norway. 
But the whole circulation tends to be less lively. Depressions are not 
so deep; winds are much less strong as a rule, shown by the fact that 





June and July have fewer gales than any other months. As a cor- 
ollary, depressions move less rapidly; not infrequendy in a bad summer 
they persist for several days in unfavourable locadons. One of 
the worst possible situations with regard to S.E. England arises 
when slow moving or stationary depressions repeatedly find their 
way into the southern North Sea (August 1915, August 1946, June 

But still the streams of tropical and polar air flow over us; slower- 
moving and less different in temperature than in winter, but still 
capable of producing frontal cloud and rain. Summer is then a season 
at which in principle at least the weather is similar to that of much of 
winter. However, the air is warmer and where saturated can hold 
much more moisture. Hence if rain falls in summer it generally falls 
more heavily although not as a rule for such long periods, as in winter. 
This is reflected by the statistics of rainfall ; the amounts in July and 
August are larger although the number of days with measurable rain 
is generally less than in January-February. Slow-moving humid air 
masses are also very liable to produce the characteristic summer 
thunderstorm, of which two main types can be recognised. The first 
of these is the series of scattered thunderstorms due to heating of a 
somewhat unstable air-mass over the land. The second is rather of 
frontal origin, and is very characteristic of the break-up of a summer 
heat-wave. So much indeed is this noticeable that Englishmen will 
still lend their amused support to Charles II 's opinion: — "The English 
summer consists of three fine days and a thunderstorm." 

Let us consider a familiar sequence. On a July day a depression 
is filling up to the East of Iceland, while pressure is high towards the 
Bay of Biscay and South West of Ireland. Cool westerly winds give 
the familiar strato-cumulus with occasional bigger cumulus and 
showers among the mountains of Scodand. In England towards the 
Midlands the weather is fair, and the warming of the surface layers of 
air in East England lowers the relative humidity; fine weather cumulus 
and a pleasant westerly breeze prevail; the cumulus tending to have 
a higher base and to decrease in amount towards Eastern England. 
Along the Channel coast the wind sets slightly off the sea during the 
day, giving stable air on the coast and long hours of sunshine. After- 
noon temperature reaches 70° on the South coast, 74 in London, 
67 at Blackpool, 62 in Skye. A clear night follows, with minimum 
temperatures down to 50 inland. 

Next day the anticyclone spreads north-eastward; winds arc still 
W. to N.W. but decrease in strength. Clear sunny weather prevails 
over most of the country ; the day is hot inland, and coastal sea-breezes 
arc well-developed. Probably 8o° is exceeded in London and die 
Thames Valley, and 76°-78° in our other Midland cities; a little small- 
sized cumulus is found here and there towards afternoon particularly 
in belts a few miles inland from the coasts. 

On the third day the anticyclone has moved so that its centre lies 
over the North Sea. In W. Scotland a gentle humid current from the 
S.W. fulfils the conditions described in the last chapter and gives 
relaxing damp warmth. But in most of England the warm Continental 
air supply with a light S.E. wind gives clear skies and a rapid rise of 
temperature to something like 85 to 87 inland in the South, and 
probably 8o° in the North. A truly hot day, with in the evening con- 
tinued warmth among the buildings of the cities; and on the succeeding 
night temperature scarcely falls below 6o° even in the country. Next 
morning the sultry air warms up rapidly; but it is probably more 
humid. For as the anticyclone retreats eastward and the south wind 
freshens the moist air begins to advance again over the country from 
the direction of France; sometimes it may even come from the Medi- 
terranean. Converging towards this warm stream of air however there 
is probably a cooler stream from the Atlantic; the two meet along a 
north-south boundary or front, which moves slowly eastward. 

At such a slow-moving front it is by no means uncommon to find 
great instability. Some of the advancing cool air is held up by friction 
at the surface, while above it moves more freely; hence over a broad 
belt of country we find that while the surface air is still hot, the air 
above is much cooler. We have already seen that under such condi- 
tions rising masses of humid air are extremely unstable; expressed 
otherwise, cumulus once it begins to form at any point grows very 
rapidly upward and may easily reach the heights at which it acquires 
the title of cumulo-nimbus, that is a height at which ice crystals form, 
shown by the fibrous appearance at the top. We have already seen 
that once this stage is reached violent thunderstorms are liable to 
break out; they develop very widely from south to north across country 
and after they have passed, the heat is dispelled by the advent of the 
pleasantly fresh and cool Atlantic air, as a rule many degrees cooler 
than that before the storm. Normal S.W. to W. conditions with 
maritime-polar air then resume their sway. 



Frontal thunderstorms of this type occasionally last for some hours 
in S.E. England, following exceptionally hot weather. Not uncom- 
monly they are associated with the front formed when cooler air lying 
over the North Sea is being overridden by hot, humid air from France. 
It has been observed that afternoon maxima over 90 in France arc 
very usual precursors of such storms. The warm humid air takes some 
hours to spread northward before reaching a sufficient height for the 
necessary instability to develop; hence in association with slow-moving 
fronts they frequently develop in the evening hours as well as the after- 
noon, and may continue after midnight, as Londoners who recall the 
great July storms of 1923 and 1945 will recall. (Gf. a paper by C. K. M. 
Douglas and J. Harding in Q. J. Roy. Met. S., 1946). London com- 
plained of a similar storm in July, 1565. 

In the main however the predominant theme is that of the west 
wind with the maritime-polar air; sometimes with occlusions and an 
hour or two of slow-moving warm rain with little wind; sometimes 
with widespread layer cloud when the south-wester of 'tropical' origin 
blows. Occasionally very unpleasandy oppressive warm moist air of 
this type gives intermittent sunshine and temperatures over 8o° with 
a wet-bulb of 70 or so in eastern England. At the same time in 
Cornwall the very low stratus is scarcely broken at all and maxima 
of the order of 66°-68° prevail. The steady forward movement 
of tropical air under these conditions gives very heavy orographic 
rain on our western mountains. A notable example occurred on 
29 July 1938 when over seven inches fell at Buttermere. (Cf. Fig. 42, 

P- '37)-. 

Yct in a dry season quite vigorous fronts may pass with pracdcally 

no rain at all in the Eastern counries. An outstanding instance befell 

on 16 September 1947. By 13I1. temperature at Cambridge was 84 

with a strong wind, a falling barometer and increasing cloud ; an hour 

later the temperature was 70 and the sky was again clear. A cold front 

of some note had passed yet very litde rain occurred. It is probable 

that this event was partly attributable to the intense drought; there 

was practically no ground moisture and even in the warm southerly 

air-stream but little cloud developed (Fig. 37, below). 

Occasional years give persistent anticyclonic drought as in 192 1; 

if the anticyclones move rather to the north, a hot dry east-wind 

summer results, as in 191 1 or 1887. A shorter spell, lasting for a week, 

of such weather marked the opening of the Olympic Games at the end 


of July 1948; on lour successive days maxima 
surpassed 90 in the South, and with the north- 
westward spread of the warm air there was a 
rare occurrence of a maximum of 90 in Scot- 
land, at Prestwick on 29 July. By contrast, other 
years give us bad summers when for several 
weeks slow-moving depressions from the AUantic 
make their way towards the Channel and on to 
the Continent. The rainy sectors of such lows 
traverse S. England, each one giving cool easterly 
to north-easterly winds and rain for many hours; 
while as the depression moves slowly onward, 
minor fronts give additional rain and cloud in 
the rear. August 191 2, 1924, 1931 and 1946 
may be mentioned as examples. In August 1945 
for two weeks persistent North Sea cloud spread 
over S.E. England in association with an equally 
persistent anticyclone in the region of the Heb- 
rides. As we might expect, this August was one of the dullest on record 
near London, while at Stornoway it was one of the finest and most 
sunny for a number of years. 

More recently, August 1 948 gave persistent heavy rains in Southern 
England, associated with a slowly-moving low off Southern Ireland 
and a very humid cloudy air stream with minor fronts within it. 
With quite a minor front between humid air from the direction of 
France and a cooler air-stream from the North Sea, very heavy rain 
fell for several hours on the night of 2-3 August; in all nearly three 
inches fell in the Cambridge district. Later in the month, a depression 
whose centre crossed England gave an appalling deluge of rain north 
of the centre. With the rainy sector lying to the north-ward we should 
expect a continuous fall for several hours in any case. But on this 
occasion the easterly surface wind off the North Sea gave exceptional 
falls on all the eastward-facing hills. Nearly three inches fell in the 
Harrogate district, and upwards of four inches in South-eastern 

Fie. 37 
1200 hrs., 16 Septem- 
ber 1947 

Plate Villa: Rays of Aurora borcalis: from Abcrticlhy, March 17, 1949. 
b: December 1946. Stormy seas breaking on the Scilly Isles. 


resemblance to the shape of cumulus will be noted, and the relatively bright light 
compared with inland at the same season. 



Scotland in a continuous downpour lasting many hours. Severe 
flooding carried away railway and road bridges, rivers rose to unpre- 
cedented heights. The results accompanying convergence of the 
surface air-streams into constricted valleys, and into estuaries such as 
the Forth, are just as evident when the wind is easterly as they arc 
when a winter southerly gale besets Snowdonia (cf. Fig. 43). The 
phenomenal Moray floods of August 1 829 on the Findhorn and Spey, 
befell under similar circumstances. There is no doubt that a series 
of active depressions crossing Southern Britain in August is one of the 
worst meteorological events that can occur in our climate. 

In general August is appreciably wetter and more cloudy than July 
throughout Britain, but especially in W. Scodand. This can in part be 
attributed to the warmer sea, and the diminished frequency with which 
the Azores anticyclone spreads far enough northward to aflect Scot- 
land and N. England. At the same lime, in association with the warmer 
sea and also, we may presume, the lack of Arctic ice, it is rather rare 
to find a well-established High to the northward. There is litde to 
impede the eastward movement of Atlantic lows, therefore, in a normal 
year, although these do not move quickly and arc not in general of 
great depth as yet. We have generally to wait till October for the 
first conspicuous results of the sharp increase of die temperature 
gradient between the rapidly-cooling Arctic and the still-warm 
Atlantic, in the shape of more active depressions and the stronger 
winds of autumn on our exposed coasts. 

In association with the warmer sea and the sfighdy less powerful 
sun, August is on the average more humid inland; the surface wind 
from the sea has a higher water-vapour content. At the same time 
with a less powerful sun the land is not quite as warm; hence we find 
August tends to be more cloudy than July. Harvest goes forward 
under varying skies, and in the north the greater proximity to the 
normal route of Icelandic lows means that there is almost always 
a good deal of intermittent rain, with consequent difficulty for the 
farmer. It is unusual, in the South at least, to experience low mini- 
mum temperatures at night, partly on account of the more humid air, 
and the dampness of the ground after rain which is rather frequent in 
a normal August. In Scotland however, the cool north wind behind a 
passing depression may give a night of clear sky and a sharp fore- 
taste of autumn before the end of the month ; occasionally tempera- 
tures below 32 ° are recorded in the latter half of August over a wide 



area. This occurred for 
example towards the end 
of August 1946 when a calm 
night of clear sky and rela- 
tively cool air befell in a 
feeble ridge of high 
pressure, behind a deep 
depression which had 
moved along the English 

September docs not in 
general continue die steady 
process of decline which 
appears to set in with the 
rather cloudy west wind of 
July, through the increased 
cloud and humidity of 
August. We have already 
seen that from time to time 
the advent of anticyclones 
of exceptional persistence 
leads to months of a very 
different type from the 
normal, the most remark- 
able example in recent years 
being August 1947. In 
September the march of 
the Atlantic lows is frc- 
quendy arrested for some 
time, and one or more spells 
of fine, quiet dry weather arc so regularly experienced in September 
that in general it is a much drier month than cither August or 
October all over Great Britain. 

Exacdy why this should occur in not yet clear. Moreover, the 
character of the month of September has tended to change; since the 
late Victorian era it has tended on the whole to be rather drier than 
previously. As it is, some Septembers arc characterised by much 
windy south-west to westerly weather, and the story of the maritime- 
polar air is repeated for yet another month. With little ice in the Arctic 

Fig. 38 
0600 hrs., 3 August 1948. Slow-moving cold 
from with very heavy rain in the Midlands 
overnight (nearly 3 inches at Cambridge) 



however the polar air is not especially cold by comparison with the 
warmer seas over which it must travel to reach us. As in other months, 
it tends to give showers in the west of the country, and to be a litde 
drier to the eastward with longer periods of sunshine. This is extremely 
important for the Scottish harvest; on the higher ground towards the 
cast the spring, as we have seen, is later. Harvest is therefore later; 
and the drying of the oat crop often owes more to evaporation in the 
September wind than to the sunshine. (See PI. XXIII6. p. 286) 

Anticyclonic clays in September arc very characteristic. The longer 
nights give more time for the temperature to fall as a result of out-going 
radiadon. Following a rainy August the ground is not uncommonly 
rather damp, moisture having penetrated to some depth. In the warm 
September sunshine by day evaporation into the air is still consider- 
able. Evaporation from ihc surface however results in more moisture 
rising from below if it is available; and thus we can sec why with quiet 
air and the lengthening nights the conditions are favourable for the 
formation of dew and mist. On a quiet evening as we have seen the 
surface air soon falls to saturation point adjacent to the ground and dew 
is deposited; and the whole process is emphasised when the surface 
layers of the soil are already damp. Hence too the Englishman's 
cautious attitude with regard to sitting out on a warm September 
evening in his garden. Occasionally a dry warm September follows 
such a dry August that the dew docs not readily form. 

It is to be observed that longer nights in themselves are not a 
sufficient explanation. The fall of temperature between sunset and 
dawn on a short June night is on the average greater than that in 
September, or indeed at any time of year unless there happens to be an 
exceptional snow-cover. This is largely because the fall of temperature 
is mainly during the three hours or so after sunset, and also the ground 
in June is in general drier. Dry ground loses heat from its surface more 
rapidly, as we have seen in the instance of sandy soils (p. 171). The air 
too is drier, allowing more radiation to escape from the ground. 

In the quiet air of September valley-inversions very readily form 
at night. In no month are the average differences of temperature more 

Plate 19 

Aberystwyth, Cardiganshire: quiet July afternoon on ihc Welsh coast. Strato- 

cumulus and cumulus in a light humid south-west air stream; cloud slighUy broken 

in the lee of Pcmbrolcohirr after passage over the land. 



Julian Huxlty 



acute between favoured lull sites and valley-bottoms inland. (Cf. 
Chap. 9, pp. 166-8). For while there is rapid cooling of the ground on a 
clear evening by radiation, this leads as we have already seen to con- 
tinuous "ponding" of the cooled air, especially in undulating country. 
No cyclist homeward bound on a fine September evening will fail to 
recognise this. 

But the air at higher levels remains warm; and the inversion 
boundary between the cool surface air and that at higher levels tends 
to subside at night, partly as the result of the slight outward movement 
of the land-breeze all round our coasts and pardy owing to the increased 
density of the cooled surface layers. 

Later in the night therefore the air on a hill finds itself losing heat 
by radiation, both from itself and from the adjacent ground. But at the 
same time the subsidence of the warmer air aloft means that inward 
radiation from the warmer air-mass above to the ground is to some 
extent balancing the outward loss. Hence we find that at higher level 
stations the temperature does not fall, later in the evening, so rapidly 
as in the valleys below. In a month such as September the upper air 
is frequently quite warm and diy, fulfilling the needful conditions. 
The minimum temperatures attained at hill stations, especially where 
die ground retains some warmth from the sun, may be many degrees 
above diose in the nearby valleys. 

Accordingly, it is in September that very marked variations in the 
incidence of early morning frost commonly occur. These particularly 
interest the amateur gardener; but they are perhaps not so economically 
significant to the fruit-grower as those of spring. The table on p. 168 
shows that over 15 years the greatest differences between the minimum 
temperatures on the flanks of the Malvern Hills and in the Severn 
valley tend to occur in September. 

With warm subsiding air aloft the stirring-up of the surface air is 
hindered; hence quiet days in September are often hazy, especially 
inland in the Midlands and South. Sometimes die light dry easterly 
drift off the continent round a large anticyclone is associated with a 
good deal of haze; rising currents carrying dust or smoke arc checked 

Plate so 

a. Vale of Ock, Somerset : July afternoon. Summer shower from heavy approach- 

ing cumulus and cumulo-nimbus. 

b. The same shower passing away. 

C.B.S. K 

Julian Huxley 



by the existence of the subsidence inversion-layer at a height of the 
order of 3,000 feet. (Cf. PI. Xlla, p. 159). We have already seen that 
the subsidence-inversion is commonly developed in large anticyclones 
detached from the region of the Azores. 

It is rare at this season to find the cold air from the Arctic 
bursting out sufficiently to build up a high-pressure region of the 
colder type such as we experience in the spring. Nevertheless, such 
things have happened; in recent experience the most memorable 
example befell on September 19-20, 1919. On the morning of the 
20th snow lay down to 800 feet or thereabouts at many places in 
Scotland and N. England, and even at Princetown (1,360 ft.) on 

In any year however, a wet windy September may occur, with 
active fronts frequently crossing the country and continuing the rain 
and wind of such an August as that of 1950. We are a very long way 
as yet from full knowledge of the factors which determine why in some 
summers, the Azores High rarely spreads towards us, and instead, 
vigorous depressions follow tracks much to the southward of normal 
expectation. The South Devon rains in autumn i960 were exceptional. 

October is a definite autumn month which in some years gives us 
quiet anticycionic spells like September; but more generally a scries of 
active Atlantic depressions with well-marked fronts begins to pre- 
dominate. There follows the normal sequence of weather, in which the 
several air-masses affect the country in greater or less degree from south 
to north, of warm dull days with maritime-tropical air, much low 
cloud and heavy orographic rain (Figs. 39, 40, below) ; of brighter west- 
wind days with maritime-polar air in its several possible varieties. 
Generally from the rapidly cooling Arctic a foretaste of winter arrives; 
occasionally we can again recognise the unstable maritime-Arctic air, 
giving the first snow, hail and sleet squalls of winter and the first 
upland snow-cover. The air flooding over the country ahead of these 
vigorous October depressions is still from a warm sea, and gives heavy 
and prolonged rain ahead of warm fronts (Figs. 39, 40) ; while polar 
air is now more unstable and prone to give showers. The effects of 
exposure to wind and rain become apparent in the woods. In the north 
of England and Scotland, leaves fall fast with the wind. It is only in 
sheltered inland valleys after an anticycionic spell that the glory of the 
autumn woods can really be seen, for example on Twccdside. But 
the South of England lies farther from the track of many lows, and 



hence the leaves generally give an attractive display of colour especially 
following the first frosts of a dry autumn. Over the greater part of the 
Midlands the first morning with a minimum below 32 in the screen 
can be expected early in October. 

November is also fairly to be described as autumn, rather than 
winter, in a normal year. Radiation fogs tend to be both more frequent 
and, with the feebler sun- 
shine, more persistent. 
Both in November and 
December many depres- 
sions tend to follow tracks 
north-eastward along our 
western coasts, so that 
cloudy days and strong 
winds or gales are fre- 
quent, with very heavy 
orographic rains in all 
our western districts. 
Inland to the eastward 
however, the wind is often 
much lighter, especially 
when pressure tends to be 
higher towards the Con- 
tinent. We have already 
seen that the lightly- 
blowing humid southerly 
air-stream under these 
circumstances is very lia- 
ble to give mist and fog, 
especially in the Midlands, 



988 /^^ J^-Or^C 

/ S y^ ^*45 
/ 904 f 


f 980 /Ck-*^ 

\ y 9ii> ^\^ . 


\ A V b&* 

\ "ndv /Pi? 

"996 ^^mjo^L-^vm? 


;f- 1000 — sr^^&-^ ©-^ 



- 1004 -~"^ ^^mt&^ 

J W? 


Fig. 30 
1800 hrs., 5 October 1943. R. rain; D, drizzle 
(after C. K. M. Douglas & J. Glasspoolc. 
Q.J. Roy. Mtt. S., 1943) 

whether through radiation at night or advection over land already 
cooled. (Cf. Fig. 4, p. 31 for 22 October 1937 and Fig. 67, p. 256 
for 27 November 1948). 

The map for 27 November 1948 is interesting, showing the situation 
during one of the most persistent London fogs for many years. Surface 
temperatures were little above freezing-point in East Anglia, but were 
nearly 20 warmer a thousand feet above the ground. The surface air 
with its fog drifted northward beneath this inversion as far as Tyneside 
and East Scotland, lifting there into low stratus cloud. 





To the westward however a 
stronger southerly breeze prevailed 
ahead of a minor cold front off 
Scotland. This air, descending over 
the mountains, became warmer still 
along the Moray Firth; and the 
noon temperature of 59 at Lossie- 
mouth accordingly contrasts very 
markedly with the dismally cold 
Midland fog. 

Somewhat rarely maritime trop- 
ical air reaches us in November to 
give temperatures over 6o°. But 
in December and January it is 
interesting to observe that practi- 
cally no example of a day-time 
maximum above 6o° has ever occur- 
red except in the lee of mountains 
when the air-stream was descending 
over them; 65 at Achnashellach 
in Ross-shire on 2 December 1948 
is the most remarkable example ol 
this. The highest January tempera- 
ture for the past hundred years in 
England was 62-5° at Durham on 
9 January 1 888. Sixty degrees has 
been recorded at Aberdeen on 
Christmas Eve (1931); 63 at Abcr in 1916, in N. Wales in the lee of 
Snowdon in January 1929, and at Rhyl in 19 16; 61 ° at Wrexham in 
January 1944; 62 in Dublin, also in 1888. Each of these instances 
affords a reminder that with a moist air current under stable conditions, 
generally round the margin of an anticyclone with warm subsiding air 
at some higher level, the air will descend on the lee side of a moun- 
tain range in much the same manner as the Swiss fohn; and our 
highest November maximum on record (71 at Prestatyn in 1946) 

Plate 21 

Stvdland Bay, Dorset: fine and nearly cloudless August evening on the south 
coast, with slight cirro-stratus hinting at a distant depression. 

Fig. 40 
Distribution of rainfall, 5 October 
1943; heavy orographic rainfall in 
warm sector air (alter C K. M. 
Douglas & J. Glasspoole. Q_.J. Roy. 
Mel. S., 1943) 

P 1. A T E 2 2 


1 10 10 6 J 2 

occurred For similar reasons. Although it can be shown that with our 
small mountain ranges the effect must be slight, it is just recognisable 
in the distribution of our highest mid-winter maxima, and perhaps 
more frequendy on the North Wales coast than elsewhere. (Cf. also 
Chapter 13). 

Severe cold in November is 
somewhat unusual; the range of 
variability of the monthly means 
(about six degrees on eidier side) is 
considerably less than in Decem- 
ber. Indeed December resembles 
January and February in the con- 
trast it can afford between a stormy 
'Atlantic' month such as diat of 
1934 (mean temperature about 7° 
above normal, Fig. 41) and that of 
1890 or 1878, when the Continental 
anticyclone built up exceptionally 
early and die mean temperature 
was 10-12° below normal over most 
of the country. It is in December 
that the sunshine duration is low- 
est, especially inland, so that the 
relative brightness of our coasts is 
then most noticeable. Plate VII, 
page 122 will bring to mind that 

even stormy December can give its own unsurpassed magnificence 
where the cliffs stand forth against the gale. 

So concludes the year; from many points of view it can be sum- 
marised as the story of the possibilities of maritime-polar air, now 
moving quickly, now stagnating; now coming over a cool sea, and 
five months later over a sea twelve or fourteen degrees warmer. In 
the far North it tends to be more unstable at all times; in the extreme 
south-west it may almost assume the stable characteristics of the humid 

Plate 22 

Huntingdonshire: July afternoon, before harvest. Cumulus clouds showing 
varying degrees of growth, but as yet litde threat of rain. Moderate westerly wind. 

afternoon maximum 74 . 

F10. 41 

Rainfall, December 1934; a very 
mild south-westerly month (from 
British Rainfall by permission ol 
Her Majesty's Stationery Office) 



tropical-air which it closely resembles after its very long sea travel. 
With relatively small overall range of mean temperature there is not 
only a wide range of possibilities. There is also a particularly noticeable 
variation in the intensity of the light from winter to summer by con- 
trast with many other populous lands. All these factors contribute to 
the diversity of our atmospheric effects and the impressions and 
recollections wc retain from our travels. 

Dr. C. E. P. Brooks has reminded us that our seasons in the British 
Isles arc only slightly accentuated, so that all through the year there 
are days which might belong to any month. And certainly the extreme 
temperatures we have mentioned for the various months go far to 
support this view. 

We have reviewed the prevailing vicissitudes in time, in an average 
year Wc may now continue to examine the variations in our experi- 
ence due to location. Small differences in wind-speed, humidity, and 
temperature arising from location affect our perceptions very markedly 
and go far to explain the determination with which many argue 
regarding the differences of climate from place to place. 

Not only the local variadons in place, but also the year-to-year 
variations in time, of the impression given by our climate, owe a very 
great deal to the extent to which the air remains in vigorous motion. 
In some months, or even for the greater part of a year as in 1949, it 
is as if die whole circulation of the atmosphere over N.W. Europe were 
slowed down, by contrast with other spells of great liveliness. Further, 
the less rapidly the air-mass moves, the greater the chance that 
modification in the direction of greater cold or greater warmth will 
take place even in our relatively small islands. Looking farther back 
in time, wc are beginning to suspect that sometimes there have been 
not merely single years, but whole centuries during which the general 
vigour of the circulation and the frequency of 'changc-of-air-mass' 
over our islands tended to be less than at present. It will be sufficient 
for the moment to point out some contrasts between the mean tem- 
perature of quiet summer and winter months, inland and by the sea. 
and the means for windy and cloudy unsetded months of the same 
name. Such contrasts, if maintained, would go far to explain the slight 
but significant changes in our past climate demanded by archaeologists, 
pahco-botanists and even some historians; and the figures below 
will suffice to remind the reader that die range of variation between 
quiet and breezy summer months is more effective inland. Some 



aspects of this variability have been discussed elsewhere by the present 

July rg2i — quiet, anticyclonic and sm-.ny: 

Inland — Nottingham mean 6G-2°, 5-2° above normal (77-9°, 54'5°). 

Coast— Skegness „ 62-8°, 3-8° above normal (70-5°, 55-0°). 

Rain 70% below normal, Sun 40% above, for East Midlands generally. 

July ig22 — cool, unsettled and cloudy: 

Inland— Nottingham mean 56-9°, 4M below normal (63-9°, 49-9°) 

Coast— Skegness „ 57-1°. 1 '9° below normal (63-5% 50-6°) 

Rain 60% above normal, Sun 25% below, for East Midlands generally. 

Figures in brackets are mean daily maxima and minima. 


Chapter 7 

Brooks, C. E. P. (1929). Climate, p. 21. London, Benn. 

Douglas, C. K. M. and Harding, J. (1946). The thunderstorm of July 

14-15, 1945. Q. J. Roy. Met. S. 72: 323-31. 
Manlf.y, G. (1951). The range of variauon of the British climate. 

Geogr. J. 117: p. 43- G8. 



. . . Humid evening, gliding o'er the sky 
in her chill progress, to the ground condensed 
the vapours throws. Where creeping waters ooze; 
Where marshes stagnate and where rivers wind, 
cluster Oie rolling fogs, and swim along 
the dusky-mantled lawn. 

Thomson : 77k Seasons 

The impression of the scenery of Britain gained by the passing 
traveller undoubtedly owes a great deal to the changing light; and 
his emotional response to any view may be totally different if it is seen 
under differing conditions of weather, even by the same individual 
on the same day. Moreover, the individual traveller may chose that 
aspect of die scenery which gives him the most satisfaction; and his 
degree of contentment not only depends on his personal feelings and 
make up. It also depends on what he wishes to sec in accordance widi 
the spirit of the time in which he lives. Hence the one time popularity 
of Monarch of the Glen studies of the Highlands, arising from early 
Victorian, and indeed Germanic, romandcism. Contrast the urbane 
and reasonable calm of the i8lh century landscape under a hazy blue 
sky; and the rcstfulness of the fashionable 20th century Hebridean 
water-colour responding to the mood of the Eriskay love lilt in which 
so many of the modern generation wish to see those withdrawn islands. 
So far we have laid the emphasis on the several types of air we may 
expect to sweep over our islands during the year. We have seen how 
these different air-masses originate; and we have noticed the effects 
arising from die tracks they follow en route for our shores. Slight 
differences are also to be noticed from season to season in the 


characteristic weather resulting from the movement of any one air-mass 
across us. In particular, that most frequent type we call maritime- 
polar tends to give showers more frequently throughout the colder 
months, although with much the same amount of cloud-cover as is 
present during the warmer half of the year. 

When the air reaches our shores the type, character and thickness 
of cloud developed owes much to the features of the landscape over 
which the air continues its course. In this chapter we shall consider 
these effects, together with those associated with varying clarity of 
the air. 

The important thing to remember about so much of the weather we 
experience is that it depends on very slight variations in die humidity 
of the surface layers of" air, up to five thousand feet or thereabouts. 
If we analyse the variations of relative humidity with height we shall 
find many occasions when, at one place, the R.H. is 95% between say 
3,500 and 4,000 feet and no cloud exists. A few miles away die air 
at the same level is just saturated at the same level and a continuous 
cloud sheet lasts for many hours. Over the year as a whole the average 
temperature at 4,000 feet over the Midlands is about 35 . Air at this 
temperature with 97% R-H. has only to be cooled o-8°F. to bring it to 
saturation. This simple illustration will serve to remind the reader 
that one of the most characteristic features of our climate is the ease 
with which low cloud develops and spreads over the sky, especially 
in hilly districts. Many of the photographs show this characteristic 
low cloud, sometimes detached (Plates 2, 5, pp. 3, 50) or more con- 
tinuous (Plates 19, XI lb, pp. 126, 159) whose formation has been 
discussed in an earlier chapter. 

But here we touch on a vital factor in our changeable British skies; 
changeable not merely in time, but also over quite short distances in 
space, as many motorists know. Our mountains rarely attain 4,000 
feet; south of the border few reach 3,000 feet; yet they play an import- 
ant role on many days and not only with regard to the formation of 
cloud in the ascending currents on the windward side. 

It is to be recalled that an inversion is said to exist when the 
temperature in a layer of air a few hundred feet thick remains steady 
or rises with height instead of falling. (Cf. Chap. 3, pp. 27, 49). If die 
cloud sheet is extensively developed to windward and the base is fairly 
low, and at the same time a well-marked inversion is present at no 
great height above the mountain summits, rising air on the one 


side is accompanied by descent on die other. This will readily be 
perceived if we consider what happens to a small mass of air following 
the streamlines over a mountain range. With an inversion layer 
above, it will have risen into a relatively warmer environment than 
itself, and so will tend to sink (cf. Fig. ga, p. 51). 

Sinking air is however warmed as it descends as a result of com- 
pression by the air above; and air which is being warmed becomes 
drier, as we have already seen. Hence it is very common to find dial 
when an extensive cloud-sheet is present on the windward side of one 
of our hill ranges, the cloud is much more broken up on the lee side, and 
may even vanish entirely. This explains why there are many days when 
a motorist crossing die Pennines or the Welsh mountains finds that the 
amount of cloud on one side is very different from that on the other. 
In a springtime anticyclone with an easterly wind Yorkshire may have 
no sunshine at all while Southport has twelve hours a day; while 
conversely, Shropshire is often sunny when Aberystwyth has grey 
skies. For such a break-up of a cloud sheet to occur it appears that 
in general the height of the top of the inversion should be not more 
than about three times that of the obstacle beneath. 

Nevertheless there are many occasions when no inversion is found 
above our hill ranges until a much greater height is reached; in such 
cases the hills play little part in the extent and thickness of the cloud. 
They may however still play an important part as regards the amount 
of rain, most of which is derived from die elevation of surface air 
currents. The diagram below illustrates the amount of rain which fell 
in an intensely humid and warm current of tropical air in July 1938; 
all over the country it was cloudy, but the rain varied very greatly 
in amount to windward and leeward of our hills. 

Rainfall varies enormously in amount over the British Isles as 
everyone knows. Moreover even among our mountain ranges local 
variations in rainfall are very considerable. The frequency and rate 
of ascent of moist air depends not only on the presence of the hills but 
on the degree to which air currents converge into certain valleys 
leading to additional forced ascent in constricted channels. Excellent 
examples arc found in the Lake District and North Wales; the rainfall 
map for Snowdonia illustrates the effect of convergence up the Glaslyn 
valley, and that for the Lake District illustrates the effect of the con- 
vergence of air-streams entering several radiating valleys on the knot 
of mountains at the head of Borrowdale and Langdale. The 'wet 


patch' round Princctown on Dartmoor lies at the head of the valleys 
running S.W. towards Plymouth. 

Fig. 42 
Rainfall, 29 July 1038; heavy orographic rainfall in warm sector 
air (from British Rainfall by permission of Her Majesty's Sta- 
tionery Office) 

Similar effects lead to a very rapid increase of rainfall up Loch 
Linnhe, and to the exceptional falls in the group of hills lying just to 
the westward in S.W. Inverness-shire. It is considered that in a small 
area at the head of Glen Garry, the average annual rainfall exceeds 
200 inches. 

The converse effect is also found. In many places horizontal 
spreading out or divergence of air currents leads to subsidence and 
hence, as we have seen, a tendency for clearer skies. The best example 
is provided by the relatively high sunshine durations in Fife and 


Angus. The divergent trend of the Grampians and the hills to south- 
ward is often very effective when the wind blows from westerly points; 
something of the same kind is recognisable in the inner Moray Firth. 


Fig. 43 

Rainfall of N. Wales 
(by courtesy of Geographical Publications Ltd.) 

On a smaller scale the combined effects of divergence and subsidence 
over mountains arc sometimes to be observed in many other districts, 
round the head of Morccambc Bay for example in north-easterly 
wcaUier, and to some extent at Plymouth. On the windward side, 
extensive stralo-cumulus or even stratus at 3,000 5,000 feet prevails; 
to leeward of the hills only occasional fragments of cloud arc to be 
seen. Any motorists driving about the north of England, Wales or 
Scotland soon becomes familiar with these effects especially in 
summer. Descent accompanied by divergence of the surface air- 
streams not only leads to decreased low cloud but also to decreased 
rainfall. Perhaps the best example of this is at the small island of 

Plate 23 

Beeley, Derbyshire: September. Fine-weather cumulus, in a westerly air-stream. 

Temperature 67 . 



Fidra in the Firth of Forth, where the lowest rainfall of any Scottish 
station is found (21-8*); similar effects obtain in the inner Moray 
Firth, at Nairn for example. 

In more unsettled showery weather the changes in the light be- 
tween flat coasts and mountain districts are especially noteworthy and 
become more marked when the sun is low. This arises partly from the 
reflection of light from the nearby sea, and also from the much greater 
thickness of the cloud building up over die mountains. 

Further, the characteristic tones of the grasses in wetter districts 
are darker. The heavy green of Molinia, which predominates in the 
hay crop in wetter districts, contrasts markedly with the lighter 
Agrostis, and above the cultivated fields dark moorland predominates. 
Limestone districts owe much of their relative brightness to the lighter 
colour of the prevailing vegetation. Among our higher hills the 
variations in the light-intensity between the sunny intervals and those 
during which the heavy shower-clouds predominate arc very much 
more noticeable than in the plains. Cloud shadows arc not only more 
frequent but more intense; this is well borne out by the contrast 
between Plates 1 and 26 (on pages 2, and 175). 

The intensity with which the gradations and contrasts of colour in 
the landscape appear to our eyes depends a great deal on the quality 
of the light reaching us. Hence contrasts are sharper when the air is 
relatively transparent, t'.*. visibility is good. The impediments to 
visibility in Britain arc of two kinds; the liquid particles in the atmos- 
phere in the form of mist, fog, drizzle or rain (to which we may add 
falling snow for convenience), and the solid particles or dust. Solid 
material in the form of excessively minute particles (of the order of 
1/1000 of a mm. in diameter) is largely swept up from the earth's 
surface into the air whenever the air is turbulent; it is obviously more 
marked when the ground is dry. In dry clear summer weather it is 
quite usual to find that the visibility decreases towards afternoon by 
reason of the diffused solid particles in the lowest 3 or 4 thousand feet — 
giving the characteristic golden haze effect of a hot summer afternoon. 
Some indeed is often carried over from the Continent, and pilots of 

Plate 24 

Buttermere, Cumberland : rain and hail shower, October. Characteristic showery 
weather in mountains with cool S.W.-VV. wind in autumn, rather unstable maritimc- 

polar-air. Cf. Plate 16&. 



aircraft arc familiar with the yellowish-brown ground haze above which 
in warm weather they commonly climb at 4,000 feet. This develop- 
ment of day-dme haze owes much to the development of instability in 
the lower layers of die atmosphere during the day. By early afternoon 
in the warmer months, as we have seen, the surface air can generally 
be expected to rise in temperature as it moves inland from the sea. 
The mean daily maximum in July at Woolacombc in N.W. Devon- 
shire is 65-7°; a short distance inland at Barnstaple 67-5°, near Exeter 
71-3°, Reading 71 -4°; it may be added that Camden Square (London) 
gives 73-5°. 

Thus many occasions will befall when the surface air is stable on 
the coast, but some miles inland it is warmed up sufficiendy to be 
unstable and to give rapidly rising thermals or convcctional currents. 
These rising bubbles of air, carried along by the wind, give rise to the 
well-known day-dme haze of warm summer weather. The longer the 
land track and the more powerful the sunshine, the greater the haze. 
For this reason fine weather with a light westerly to northerly wind is 
commonly marked by excellent visibility over the sea and near the 
west and north coasts; but while from Skye one may see the blue 
Outer Islands all day long, in Cambridge, on the same day and in the 
same air-mass it will probably be difficult to see any object more than 
three or four miles distant. 

If however, with a light south-west to westerly wind in the warm 
months the sky is overcast the visibility even far inland often remains 
very good ; turbulence in the surface layers is checked. Driving up the 
Wading Street beyond Adierstone die hills of Charnwood Forest 
stand out clearly; from Dunstable Downs the crests of the Cotswolds 
are seen. Further north, where low cloud sheets are more often found 
one is frequently reminded of Belloc's lines: 

" The men who live m the North Country 
I saw litem for a day 
Their hearts are set on the waste fells 

Their skies are fast and grey. 
From their castle walls a man may see 
The mountains, far away," 

Stand on the batdements of York and see under the pale grey strato- 
cumulus of a west-wind September noon the long low line of the 
Pennine Hills to the north-westward; or recall the days when from 
Shrewsbury the Welsh border was watched for signs of movement 


under a similar sky. Not that East Anglia is free from such grey 
skies in early autumn, as Wordsworth says of the way to Cambridge 
in 1789: 

"// was a dreary morning tuhen the wheels 
Rolled over a wide plain o'er hung with clouds, 
And nothing cheered our way till first we saw 
The long roofed chapel of King's College. . . ." 

Generally speaking visibility is better when after rainfall the air is 
subsiding and convection is limited; hence as a rule it is excellent in a 
narrow wedge of high pressure between two passing depressions. 
Further, when the weather is warm and sunny inland sea-breeze 
phenomena become well marked on our coasts. The existence of a sea • 
breeze implies that some slight descent of air is proceeding just offshore. 
Not only is the air relatively free from impurities but the sky in a strip 
along the coasts tends to be rather freer from cloud ; hence in summer 
it is commonly found that the visibility is better and the sunshine 
duration slightly higher on the coast, than inland. Over the year as 
a whole, for example, Lowestoft averages 1,720 hours of bright sun- 
shine, Norwich 1,600. Any visitor familiar with the weather at a coast 
resort will not fail to have noticed, too, how hazy the air is on occasions 
when in summer the wind continues to blow from die land. Nowhere 
is tliis more noticeable than in Lancashire. The occasional day of 
stilling, brassy heat at Blackpool accompanied by a soudi-east wind 
from the smoky inland towns and an afternoon temperature in the 
eighties provides for many a disappointing contrast with the more 
normal fresh westerly breeze, bright sky and an afternoon temperature 
of 66°-68° in August. 

It may be asked why on such a warm day the surface sea breeze does 
not set in and undercut the hot wind off the land, inasmuch as the 
sea is still cool. But if the air at greater heights is sdll very warm, as it 
sometimes is when a deep stream of warm air spreads from the con- 
tinent, normal sea breeze circulation across the coast cannot establish 
itself. Any cool air spreading inland from the sea cannot rise from the 
surface as the air above is too warm by comparison; hence no cir- 
culation is set up and the heat is equally great on the coast or inland. 
The temperature of 93 at Bourncmoudi on 16 August 1947 offers a 
good example; also 74 at Cromer in early March beside the cold 
North Sea (p. 88). 



Indeed the brazen sky now and then in summer accompanying the 
hot south-easter at Blackpool, or for that matter the north-cast wind 
of Central European origin at Brighton is a reminder that to the 
natural dust in Britain's atmosphere, derived by normal processes 
from the earth's surface, we must add the man-made smoke. Dust 
haze and smoke are far more effective than most town-dwellers realise, 
in cutting down the quality of the light and detracting from the colour 
of the landscape. A well-known artist at Lamorna Cove once described 
to the writer how he had one summer been invited to paint in Derby- 
shire; "but", he said, "when I got there the atmosphere was yellow. 
There's no light for painting at all." This was said on a Cornish April 
day when he had stopped work for similar reasons— because of the 
north-east wind— the rare wind which, turbulent due to ground 
heating on a spring day and travelling over a long stretch of land to 
West Cornwall, brings the haze which takes almost all colour out of the 
landscape even in the far west. In western Ireland the wind from the 
east is known as 'the black wind'; though it is probable that this refers 
to other evils beside diminished visibility. 

But all over England and Wales, over part of Scotland and even in 
Ireland, the effects of the man-made smoke are at times apparent. On 
many days when the Yorkshireman of a former day looked out from his 
castle walls in the clear air under a silver-grey sheet of slightly rippled 
stratus, he now finds an undoubted brownish cast; and this is equally 
true of Lincoln. In the Lake District, even the Glasgow smoke reaches 
Skiddaw on occasion; and in particular the south wind ahead of a 
slow-moving warm front often brings an unplcasing additional gloom 
from Lancashire.' In Cambridge the smoke of London is at times 
perceptible under similar circumstances and it has been known to 
affect visibility at Norwich. For the existence of extensive layer cloud 

' Compare here Wordsworth's poem "written on the mountain Black Combe" 
the gloom beset a "geographic labourer", Col. Mud S c of the Ordnance Survey, 
in 1013. " 

Plate IXa: Daffodils in early May in die high valleys of die Scottish border. 

Characteristic cumulo-stratus associated with anticyclonic weather and a light 

N.E. wind towards the North Sea coasts. 

b: Near Baildon in the West Riding of Yorkshire. The fine weather 
cumulus ofa fine day in early summer; the turbulent westerly breeze of the Peanines 
shown by smoke; rather dark-coloured quarries and grey stone by contrast with 

Plate XlXa, p. 234. 

Plate lXa. 

The Timet 

C. H. Wood 

Thomas H. Elwell, Mew Turk 

Tlit Times 


frequently implies that just above there is an inversion of temperature 
through which the rising air from the cities cannot ascend ; hence the 
smoke may be borne at a higher level many miles from its source. 
Some years ago a well-known meteorologist plotted the visibility at 
various places during a motor ran across Central Scotland ; he showed 
how the Glasgow smoke haze spread eastward along the foot of the 
Ochils and seriously affected visibility on the opposite coast sixty miles 
away. With light winds the smoke from industrial areas sometimes 
follows quite a narrow track. The writer has stood on a nine hundred- 
foot summit in Durham in a light north wind, with the Cheviot clearly 
visible to the north fifty miles distant, yet, due east, the visibility was 
less than two miles — the result of the smoke haze drifting off Tyncside 
and the Durham coalfield. Pilots will remember the difficulty with a 
similar north wind and splendid visibility elsewhere, of landing on 
West Lancashire air-fields in winter when smoke haze from Preston 
was obscuring the ground below. 

Smoke is indeed a gloomy subject; for while there are those who 
will rhapsodise over the hazy City sunsets across the Thames, there are 
many more who deplore the vanished glories of a sunny little Man- 
chester from which the green and gold Pennine slopes were visible on 
many April mornings two centuries ago. It is possible that the whole 
country is affected more than we think. Early last century the Ord- 
nance Survey sighted the Welsh mountains from Bardon Hill in 
Leicestershire. Ralph Thorcsby espied the shipping in the Thames as 
he rode over Harrow Hill in May 1702. Celia Fiennes saw the Isle of 
Man from near Chester; Defoe (1726) was informed that from the 
Cheviot the view extended to the Tyne, and George Smith saw the 
cliffs beside the North Sea from Crossfell in 1747. But Wordsworth 
noted the London smoke haze from Hampstcad Heath. 

In many parts of Britain smoke from the towns spreads in certain 
well marked directions. Studies of the frequency of surface wind in 

Plate Xa: Unsettled summer afternoon over the Gareloch looking towards Dun- 
bartonshire, June 1945. Humid westerly wind; extensive low cloud breaking a 
litUc in die lee of the hills. Cumulus beginning to tower above the low cloud, 
threatening showers with the possibility of thunder. 

b: Harvest in County Durham, looking south-eastward from Plawsworth 
to the hills of East Durham. August afternoon with characteristic tendency to 
unsettled weather supported by the distant cumulus with a hint of cumulo-nimbus 

over N.E. Yorkshire. Clouds show signs ol flattening out over the sea. 

T..B.S. L 


quiet weather have been made which point to a tendency for the wind 
on many winter nights to set slightly from a southerly point near the 
flanks of the Cheshire Pennines. Perhaps this helps to explain why in 
more recent years the development of the suburbs of Manchester has 
taken place to the soudiward where the air is cleaner and purer; in 
an earlier generation it was more fashionable to live on the rising 
ground to the north and north-west. Generally speaking the surface 
relief influences to an appreciable extent the movement of air whenever 
the wind resulting from the pressure gradient, i.e. the spacing of the 
isobars, is less than force 3 or about 10 miles an hour. Canalising of 
the flow of air along valleys and the like then begins to become notable 
as regards the spread of smoke haze, especially at night and in die 
early morning when there is commonly an inversion a few hundred 
feet above the surface so that the smoke cannot rise. 

It is important to recognise that over uniform country smoke, 
especially in the day-time, is distributed upwards as well as sideways 
by turbulent eddy motion in the wind. Hence in favourable circum- 
stances wind can quickly remove nine-tenths of the smoke from surface 
air. Recent studies of atmospheric pollution in Leicester show that in 
practically every type of weather the highest concentration of smoke at 
street level was in the centre of the city, but that most daylight is cut off 
between a half and one mile down wind from the centre. There is also a 
horizontal spread with distance, but in more hilly country this may, as we 
have seen, be restricted by the topography. Details are given in a recent 
publication of the Department of Scientific and Industrial Research. 

It is evident that if unpleasant smoke and fumes arc to be liberated 
this should be done in the middle of the day when in general turbulence 
is most active. At night turbulence in the surface layers is often 
checked by the development of inversions; in such an event the smoke 
cannot be carried upwards. Accordingly, a light drift of air is sufficient 
to affect suburbs to die leeward. Dwellers in nordi-west London were 
particularly affected by this during the prolonged spell of fog at the end 
of November 1948, accompanied as it was by a light south-cast wind (Fig. 
67, p. 256). Farther up the country the conditions were often worse; and 
the unpleasant acrid murk in Wharfedale on that occasion will long be 
remembered as the penalty of living to leeward of the great Yorkshire 
coalfield and the belt of large towns on its north-western edge. 

Smell plays its part in our impressions and one might fairly 
remind the airmen of the West Riding Squadron during the war of the 


proverbial saying diat in a fight west wind the pilots could smell their 
way home all the way from the Yorkshire coast. In Hampshire Gilbert 
White himself recognised the haze "with somewhat the smell of coal 
smoke" at Selborne, and that such a "blue mist" was always accom- 
panied by a north-cast wind. For the meteorological conditions 
affecting the olfactory powers of such animals as deer, reference may 
be made to Dr. Fraser Darling's work in this series on the Natural 
History of the Highlands. 

A good scenting November morning in Leicestershire owes much to 
the combination of damp clay, humid air and little turbulence associ- 
ated with the soft grey cloud sheet of a typical quiet autumn day on 
the flank of the Continental anticyclone, when a distant Icelandic 
low brings the gentle south-wester across England. The melancholy 
smell of cabbages on dull afternoons in early December on the Gault 
clay outside Cambridge will also be recalled; similar phenomena are 
no doubt equally familiar to Oxford men, as they are throughout the 
clayey Midlands and can almost be described as an integral part of our 
own impressions, if not of the scenery. 

Generally speaking, polar air gives the best visibility though much 
depends on the length of its land travel. Towards the north-west High- 
lands and islands the frequency of excellent visibility in the pure un- 
polluted air with a long sea travel is greater than in any other part of 
Britain. In 1878 a Scottish amateur meteorologist recorded that over 
twenty-one years Lochnagar (45 miles distant) had been visible on an 
average of one day out of four from Aberdeen. The deep blue distances 
of a September afternoon in Argyllshire provide the traveller with yet 
another enlivening contrast when he recalls the hazy golden light over 
the stubble of Norfolk in the same air-mass. The characteristic bluc- 
ness of the more distant Scottish hills, especially in the Highlands, has 
been remarked by all travellers. Indeed the very name Cairngorm 
means 'blue hill'. Seen from die great Inverness-Perth road on the 
long descent to Carrbridgc, from the hills of AfTric or from die outskirts 
of Aberdeen the appropriateness of the name is very evident through- 
out the warmer months. 

The visible solar spectrum is composed of light of different wave- 
lengths from red through green and blue to violet. As it traverses our 
atmosphere the shorter wave lengths are scattered in all directions by 
the molecules of which it is composed, and by the excessively minute 
particles of dust which float in it. In a pure atmosphere free from dust 


the scattering is almost all of the blue and violet, so that for example we 
perceive the colour of a clear sky to be blue; and at high altitudes the 
sky becomes noticeably darker as less light is scattered. It may be 
added that if the larger particles we call dust are present — even the) 
are so small that dierc may be many thousands in a cubic inch of air — 
the scattering applies to all the light traversing. An atmosphere with a 
good deal of dust in it appears much more white. Hence the sky be- 
comes whiter near the horizon, pardy because one is looking through 
a greater thickness of air, pardy because in the layers close to the 
earth's surface there is more dust. 

Now if the air is clear and free from dust the scattered light, giving 
the same impression of blueness, is as it were superimposed by the 
atmosphere between the observer and the hills. Indeed we find that 
we even tend to esdmatc distance by the blueness of the distant hills. 
Hence the characteristic blue distances of the Highlands in clear 
weather are principally to be attributed to the purity of the air; in 
general this has had a very long travel over the Atlandc, and then over 
a country in which little dust is raised, partly as a result of die relatively 
frequent rains. Prolonged dry weather, especially if die wind is south- 
easterly with a long fetch over the land produces a characterisdc haze 
in which objects even at a mere five miles distance appear almost 
colourless. In Skye we are then reminded of the similar effects 
seen on sunny days in Cornwall with a north-easter, mendoned on 
page 142. 

Optical effects in regard to landscape have been the subject of a 
well known work by Minnaert. In this country Mr. James Paton, of 
the University of Edinburgh, is a meteorologist and physicist who has 
recently given a most attractive account of some of the less known 
effects; the above paragraphs arc largely based on his article. 

Topographical Effects on Wind, and the Results 

We have already indicated some of die more obvious results of the 
canalisation of the movement of surface air-streams by hills and other 
features. This becomes especially noteworthy where a broad water 
surface, as in die Firth of Forth, gradually increases in width between 
the hills. Friction is less over the water, and whenever the isobars 
allow for the development of a wind from between south-west and 
north- west the set of the wind in the Forth tends to be from westerly 


points and the strength is a little greater than it would be if there were 
no hills. Buildings and other obstacles play their local part; Scotswomen 
may be reminded of the frequency with which the wind hurries 
the shoppers in one direction or the odicr along Edinburgh's Princes 
Street on breezy winter afternoons. Similar effects are found in a 
great many estuaries and channels. The pleasures of yachting off 
Cowes are enhanced by the canalisation of the sea-breeze up the 
Solent on a warm August afternoon. 






1'in. 44 

The Solent and Spithcad. Tolland Bay (p. 90) is due south of Hurst Castle 

It has already been shown that the wind tends to descend hill 
slopes more forcibly if an inversion-layer is present above the summit 
at some height which should not be too great; one may say, roughly 


three times the height of the obstacle surmounted. The best example 
of this process in Britain is provided by the 'helm wind' of Crossfell 
and the neighbouring Pennine slopes in Cumberland and Westmor- 
land. On days when the north-cast wind prevails it is often found to 
blow with exceptional strength down the treeless slopes of the Crossfell 
escarpment facing the Eden valley, and over the fields and villages 
lying at the fool. A little way farther to the south-west, however, the 
air is nearly calm. Accompanying the strong wind on such occasions, 
a characteristic wall of cloud is seen lying along, or just above, the 
summits of the Pennincs. Its top is generally smooth in outline. 
Parallel to this cloud, known as the 'helm' — a word probably (though 
not quite certainly) expressive of die helmet-like nature of the cloud — 
there lies, at the same level but four or five miles distant to the south- 
west, another line of clouds sometimes continuous, sometimes broken 
up into fragments. This is called die 'bar'; it is seen as a cloud 
stationary with respect to die ground but in vigorous motion within 
itself. The phenomena are illustrated in Fig. 45 below. 


BELOW 6000 FT -> - BAR 






3-8 MILES 

Fic. 45 

Norma) helm wind; bar formed at crest of 'standing wave' to leeward of the 


It is evident that the escarpment lying transverse to the. surface 
wind acts like a submerged weir in a stream of water; a standing wave 
is set up just below the weir. High above this surface wave, stationary 
lenticular clouds sometimes appear (PI. XXIi, p. 270). The clear sky 
between helm and bar is associated with the descending air; where the 


air again reaches the same level as the helm, cloud is again present as 
the bar. This bar is continually forming in the rising current on its 
eastern side, and at the same time dissipating as the air descends on the 
west. It often shows rotary motion, a horizontal eddy being formed. 
Various modifications and partial developments occur which go to 
show that the upper edge of the helm, marked by an inversion, should 
not be more than 6,000 feet above the sea, the average height of the 
range being 2,500 feet, falling 2,000 feet to the valley. Moreover the 
phenomenon does not develop unless the prevailing strength of the 
north-easter at say, Tyncmouth exceeds about 15 m.p.h. It is also 

Fig. 46 

13 hrs., 29 January 1939 and 18 hrs., 17 May 1939. Moderate to fresh E. to 

N.E. wind across N. England; situations giving rise to helm wind on the 

Crossfell escarpment (For notation, see p. 5) 

evident that if the westward slopes of the escarpment were much steeper 
the How would not be smooth, and places near the foot of such a steep 
slope would intcrmittendy experience severe gusts. Something of this 
kind is known to occur in the outer harbour when an east wind (die 
'Levanter') blows at Gibraltar, where the steep-sided Rock presents 
a north-to-south barrier 1,400 feet high and direcdy in the path of 
the wind. 

Now if we extend these ideas more widely it will be evident that 
many places among the mountainous Scottish coasts will be subject 



to variable squalls descending over the crests and giving disturbed 
patches of water, on occasions when the wind is strong and especially 
when the summits are capped by cloud. Much depends on the length 
and steepness of the ridge over which the wind blows, as well as the 
prevailing strength of the wind. If the wind is only light or moderate, 
up to ten or twelve miles an hour, it seems that as a general rule the 
flow of the air would tend to become canalised along the foot of the 
ranges rather than across their crests. Many squally and disturbed 
patches of water, many windy and gusty corners among our valleys 
and hills, can be found which owe much to the surrounding topo- 
graphical features. At least one authority has suggested that the trail 
in the water attributed to the Loch Ness Monster may well be ascribed 
to the localised flaws on the surface caused by sudden descending gusts 
at a time when the water is otherwise quiet. In the eighteenth 
century it appears that a similar phenomenon, locally called a 'bottom 
wind', was described in Derwentwater. Phenomena analogous to the 
helm wind have been reported in Britain from the slopes of the 
Grampians in Angus; here, as might be expectcd,the general wind was 
north-westerly. Something of the kind occurs at times to the west of the 
Derbyshire Pennines and no doubt other examples will be found. The 
possibilities of lift they provide are of the utmost interest to all sail- 
plane pilots while the upward extension of such surface effects into 
the higher atmosphere provides an elegant problem for mathematicians, 
as Dr. R. S. Scorer has lately shown. 

Hill slopes rising steeply in face of the prevailing wind are often 
particularly windy towards their crests, and the uprush of air is useful 
for purposes of soaring flight. Where the valley below is well warmed, 
a thermal effect may be added to that of the prevailing wind; at 
gliding sites such as the Dunstable Downs and Bradwcll Edge in die 
Peak District of Derbyshire much use is made of this in summer. Bui 
all these local variations in wind speed play an additional part in 
moulding our scenery. Quite apart from questions of soil, it becomes 
very difficult to establish trees in windy sites; animals and men too 
often avoid die wind, or save their energies by adopting the prone 
position. Hence the establishment of many of our villages in just those 
locations where in conjunction with other needs, greater shelter was 
afforded; Telscombc, near the crest of the South Downs between 
Brighton and Newhaven, offers a good example, in contrast to what 
many deem the breezy exposure of Peacehaven nearby. 

10 o 10 



Hence, too, still another factor leading to diversity of scenery. 
Nowhere in England is the effect of wind better seen than on the north- 
east coast. The bare cliff-tops of Durham and Cleveland, so well 
known to those who visit Whitby Abbey or the bleak ugliness of Black- 
hall Colliery, contrast emphatically with the thickly-wooded denes, 
with Brignall Banks in 
Teesdale celebrated by 
Scott, or with the wood- 
lands of Eskdalc, where 
the Esk has cut so deeply 
into the North York 
Moors. In Banffshire die 
north-cast corner is 
nearly treeless; but inland 
the splendid woods of 
Glenlivet at 500-800 feet 
give emphatic evidence of 
the value of shelter. For 
centuries our travellers 
have been even more struck 
by the contrast between 
the treeless windy uplands 
of Cornwall and the lush 
vegetation wherever there 
is protection, especially 
along the inlets on the 

0900 lirs. 4 various 
58 yearc pbjei '..iti.. ui 

35 years obiei vatiens 

13 hrs. 
10yr*. 1926-29, 1931-36 

44ycars observation J 

Fig. 47 
Observations of wind direction; means of 
the monthly percentages of observation from 
each direction made at various coastal points 

south coast such as the Helford river; and in South Devon, Dartmouth 
and Salcombe are similarly favoured. 

Towards the British coasts we may often recognise in the distortion 
of the trees the effect of the dominant wind from the sea, as distinct 
from the more prevalent south-wester. This becomes very evident on 
the Northumberland coast, and even in the suburbs of Sunderland 
the ash trees arc woefully stunted through the combined effect of wind 
and cool air. W. V. Lewis has distinguished the importance of the 
dominant north-east wind with a long fetch over the sea in giving rise 
to powerful waves whose onset has done much to mould the features 
of the North Norfolk coast and elsewhere, although it blows far less 
frequendy than the south-wester. It seems probable that our impres- 
sions of coastal landscapes are also affected by our subconscious 

'5 2 


recognition of the dominance of the gales from the sea. The sight of 
the bent trees along the Lincolnshire marshland and the landmark of 
Wintcrton Church tower north of Yarmouth brings to mind the stories 
of the Cromer lifeboat, and for how long the north-easter was justly 
dreaded by the Newcastle colliers on the way to and from London if 
it caught them on the long stretch between Wintcrton Ness and 
Flamborough Head. Defoe mentions how in one year (1692) two 
hundred ships were wrecked in a single storm. 



a: — 

■- .1 .UulU- 


_ io»M.nnfi«6'm 

E. T 






II Ii j 

— ' — — wiy — — 






; ' 

kl* 1 i , 

:a "1 


• ■ 

1 1 Vm '™ 

■«■■■&. H^di " JL _ 



r j1 

r* ^\*\m '■ 


i, r^z ~~ 

- - - 

it— ^ 

-r-r ■ y i - 

. -i ! ! 

-H * 1 i 1 : ► — I 

— r==- 

"-p ri t >" r "V ' r*t*"'" T"* i "" fT'i n n r 



Fto. 48 
Highest gust at Tirce. 28-29 January 1927. 108 m.p.h. 

Lee slopes as wc have seen can receive steady winds, strong gusts 
at intervals, or no wind at all, depending on the overall force of the 
wind and on the degree of steepness of the slope. Sometimes back 
eddies develop (Fig. 9) and wc can well understand now the develop- 
ment of the rapidly varying swirling masses of cloud among the irreg- 
ular passes, peaks and corries of our heavily glaciated mountains. 
These effects have become familiar in the work of many artists. English 
and Scottish painters in particular have been perennially fascinated 
with the expression of the lively and complex rhythm of the movement 
of the air amid the mountains as well as that of the sea along our 
rock-bound coasts. Even our instrumental records indicate at times a 
rhythmic fluctuation in wind speed, illustrated for Bell Rock in the 
Firth of Tay by Mr. Ernest Gold in a Presidential Address ("Wind 
in Britain") to the Royal Meteorological Society in 193b. 


It would be appropriate to add a note on the occasional extremes 
of wind experienced inland. Great gales on the coast blow with a 
higher average speed, as die anemograph trace associated with such a 
gale at Tiree shows. On this occasion the highest gust reached 108 
m.p.h., while for one hour the average speed of the wind was 6G m.p.h. 
(force 11). 

Inland in a great city a really windy 
day typically gives a result such as that at 
Kensington. Although the highest gust 
reached 70 m.p.h. the average wind speed 40 V- 
for a whole hour was not greater than 
force 6 (about 28 m.p.h.). 

By reason of surface friction the mean 
speed is much lower, but occasional gusts 
still reach, or nearly reach, the violence 
and strength of those at sea. It is these 
gusts which do the structural and other 
damage. For the growth of our largest 
trees some degree of shelter is generally 
necessary, such as thai provided by an 
extensive surrounding stand of well 
grown forest or sometimes by neighbouring 


Severe gales are of more frequent occurrence during the months 
when most of the trees are bare, as Table X in the Appendix will 


Gustincss for the reason mentioned above is much more marked 
inland than at sea; the winds over open plains and uplands are also 
less gusty dian in wooded and cultivated lowlands, but arc still not 
so steady as at sea. For a second factor enters into the development of 
gusty winds. Under conditions of cool air flowing over ground which 
is being rapidly warmed, for example on a sunny spring afternoon, the 
air near the ground is very unstable; warm bubbles rise on the least 
provocation and arc replaced by chilly gusts, which feel the colder as 
the air under such conditions is generally rather dry. Hence a most 
characteristic type of day especially in East Anglia; the bright warm 
sun, especially hot in sheltered corners, but die gusty chilly north- 
easter finding its way round the houses and copses at intervals, and 
giving rise to appreciable soil erosion in the Fens after dry weather. 

Fie. 49 

South Kensington : a strong, 
very gusty westerly wind. 
1 OA 12 January to 2/1 13 
January. 1930. Highest gust 
at 70 m.p.h. 



Here the black peaty soil is absorbent of radiation while at the same 
time it is very fine; hence great turbulence develops with the strong 
heating of the ground, sweeping up the fine material. It will be noticed 
that the worst damage occurred in late April, 1943 — the same season 
of year when, for similar reasons, the worst dust storms of the Middle 
West have occurred. 








idLih ,Ln i 





Fio. 50 
Anemograms from St. Ann's Head. Pembroke, and Cardington, Bed- 
fordshire, 5-6 June 1944, from oh. to gh. "D-day"; strong westerly 
winds, decreasing in the night (by courtesy of the Director ol the 
Meteorological Office) 

One may comment on the statement often made by Southern 
visitors, how few large trees are found north of a fine from Mersey 
to Humber, except in particularly sheltered places. As the climatic 
differences in respect of temperature and sunshine are small it seems 
probable that this is partly a matter of wind, partly of less fertile soils, 
for it is indeed true that in the aggregate the North is more breezy and 
more cloudy, especially in summer. The impression is probably 
enhanced by the fact that so many ol the most frequently visited parts 


of the North, whether for business or pleasure, lie in the uplands. 
Moreover, in counties such as Durham the clearance of woodlands in 
the past was particularly active, notably in the eighteenth century. The 
effect of soil however must not be forgotten. Quite sizeable beeches 
are found on the limestone near Penrith a thousand feet above sea- 
level; nearby, other soils arc nearly barren of timber. Of the various 
trees so characteristic of the English scene the oak is perhaps the most 
celebrated, but farther north on colder soils and on higher ground 
it is the ash which attracts the attention. 

"The oak and the ash, and the bonny ivy-tree" sang the homesick 
north-country maid of Elizabeth's day; and the ancestral stock of 
many an accomplished English family is derived from a stricdy-built 
hill farm in a wide and windy country of stone walls, whose straight 
and sombre lines in autumn were only interrupted by the struggling 
ash trees and the hurrying cloud above. Of the Welsh border country 
and even of parts of high Leicestershire much the same may be said. 

Far down to the westward in kindlier Devon and Pembroke, in the 
Lleyn, the Isle of Man and even in West Cumberland, the high turf 
banks dividing the small irregular fields are most noticeable features 
of the scenery. Climate again has played its part; in the damper west, 
the keeping of animals has always been more prominent; pasturage 
and hay are more abundant. Both grass and stock benefit from shelter, 
and in a country where hedges and trees are less easily established the 
turf-grown banks with their wealth of wild flowers have survived from 
time immemorial. The extended enclosure of the open Northern hills 
by means of stone walls was a later matter; much of it began in the 
seventeenth century, and continued thenceforth until after 1800. 
Readers may make whatever they wish of the fact that they were 
largely contemporary with the evolution of Cartesian geometry, the 
Fahrenheit thermometer and the well-tempered clavier. The erection 
of stone walls rather than turf banks has indeed an extremely con- 
spicuous effect on the landscape, but interpretations in terms of climate 
would beg too many questions to be essayed here. That die greater 
windincss and wetness of the north and west, the uplands and exposed 
seacoasts has influenced in numerous ways the characteristics of the 
country that we now sec is evident to us all. The immediate remark of a 
Bradford sixth-form girl returning from a first term at a University in 
the South of England was "how much more colour there is"; but grit- 
stone and smoke, laburnum and brick play their part as well as the 



climatic factors and ihe prevailing northerly aspect of her birthplace. 
(Plates 14, IXb, pp. 99, 142). 

Thermal Effects: Sea and Mountain Breezes 

Whenever there is an appreciable difference in the surface tempera- 
ture across a sharp boundary local air movements are set up. English 
houses become far more draughty in winter, when the temperature 
difference on either side of ill-fitting windows is large. By far the most 
conspicuous of these thermal effects are the sea-breezes resulting from 
the sharp difference of temperature set up across a coastline in warm 
sunny weather. The land being so much warmer than the sea a shallow- 
local flow of air sets in almost at right angles to the coast if the weather 
is calm; typically on a hot summer day it begins to be noticeable as a 
light breeze about 10 a.m. and may continue till 6 p.m. or so. It is 
generally strongest, and penetrates farthest inland, in mid-afternoon. 
The opposite effect — the surface flow of cool land air towards the 
warmer sea — occurs in the later hours of the night towards dawn. A 
not uncommon feature of the British scene, therefore, was the setting 
forth of the local fishermen in the earliest dawn of a summer morning, 
returning with the fish soon after 10 for quick sale to the diligent sea- 
side housewife ; something of this activity is indeed still found. 

Sea breezes of this kind arc quite shallow, about 500 feet being 
common. This is easily to be observed on some of our industrialised 
stretches of coast, as the diagram below will show. On that summer 
day the afternoon maximum at Durham was 73 with a west wind, 
force 3; in an open car the sea breeze began to be felt, undercutting 
the west wind, about two miles inland from Scaton Carew just south 
of Hartlepool with its smoking chimneys; and on the beach the 
temperature was 58 . It will be evident from the diagram that the 
inflowing air from the cool North Sea attained a temperature 
corresponding with that of the prevailing westerly wind about two 
miles inland. About 6 p.m. the smoke of the chimneys was no longci 
being carried inland, but rose and spread for a time rather erratically; 
by 7 p.m. the steady drift of smoke seaward was noticeable. (Sec Fig. 51). 
It is evident that the sea breeze often sets in towards the coast from some 
miles to seaward. Watchers of the shipping from the cliffs above 
Scarborough and Whitby will often observe how the smoke of the 
steamers making their way between Forth or Tync ports and London 


spreads out horizontally some hundreds of feet above the ship, hanging 
in persistent wreaths at that level which indicates the upper limit of 
the cool sea air. 

Sea breezes however, sometimes combine to a greater or less degree 
with whatever prevailing wind exists as a consequence of the overall 
trend of the isobars on a given day. 



n ou n i r » ggeaoT 

Fig. 51 

A July sea breeze at Hartlepool on the Durham coast 
a. Mid-afternoon; depth or sea breeze on coast about 500 ft. Westerly wind 
inland at Durham and overlying sea breeze, shown by movement of smoke. 
Sketched in 1935. b. Early evening; sea breeze ceases, westerly gradient 

wind prevails 

Moreover, sea breezes do not invariably develop or spread far 
inland in the hottest weather. In order that the sea breeze may spread 
inland it is necessary for the warmed air to be able to rise, and this will 
take place more freely if the lapse-rate is large; that is, if the air falls 
off in temperature fairly rapidly with height. When the weather is 
very hot and clear the air even at 2,000 feet may still be at a tempera- 
ture of 8o c . In order that any surface air flowing from the sea can rise 
freely it will have to attain a temperature of 91° or more at sea level; 
the rate of fall of temperature with height will then just exceed the 
"dry adiabatic", 5'4°F. per 1,000 feet. Hence we find that on 
occasions when a deep warm current of air is spreading from the 
Continent, the temperature in the streets of Brighton or Eastbourne 
can be just as high as in London. The baking crowds beloved by the 
newspaper photographers roast themselves in the glare as close to the 
cool water as possible, to catch whatever feeble puffs of air they can. 
Since observations began, the highest values for upper air temperatures 
at South Farnborough in such air-masses lie in the neighbourhood of 
70 at 5,000 feet; such values imply that in theory the afternoon 
temperatures at ground level might attain 97 even at the seaside, a 
figure well in keeping with experience (93° at Bournemouth, August 
1947; 92° at Margate, August 1932; 91° at Southport, July 1948). 


The varying impression of bracing or relaxing qualities associated 
with our seaside resorts probably owes much to the extent and char- 
acter of the daily exchange of air between sea and land in quieter 
weather, especially in summer. Throughout many summer days the 
prevailing pressure gradient favours westerly winds with an Azores 
high slightly to the south west, an Icelandic low moving cast towards 
Norway. Now if we assume that the thermal effect operates at right 
angles to the coast it will readily be seen that at Brighton, for example, 
thermal and gradient winds combine on many occasions to give a 
resultant from W.S.W. ventilating the greater part of the town. 
Farther along the coast at Eastbourne the gradient wind from W. is 
slightly held up by friction over the land and the combination of 
gradient and thermal wind is not so strong, and presumably is felt 
over a smaller area during fewer hours of the day. At Bournemouth 
an effect similar to that at Eastbourne appears to arise from the shelter 
provided by the Isle of Purbeck and is probably more marked; and 
the net result of the relatively diminished vigour of the movement 
of air from the sea may be that which leads many to consider Bourne- 
mouth as 'relaxing and sleepy', while others appreciate its 'pleasantly 
sheltered' qualities. There is no doubt that the local complexities of 
air movement and exchange along our sea coasts deserve further 
study; no precise assessment of the qualities which distinguish the 
air of our seaside is yet possible. The distinction between Hoylake and 
West Kirby, Margate and Ramsgate, St. Ives and Penzance can 
however be reasonably attributed to the varying frequency and vigour 
with which the sea air undercuts the air off the land at each place 
in quiet weather. There are many naval meteorological officers 
accustomed to dealing with the problems of particular harbours and 
coastal airfields who will readily provide useful explanations of numer- 
ous happenings at our holiday resorts. 

The general regard in which the bracing qualities of our east coast 
are held owes much to the sea breeze from the cool North Sea. 
Accordingly, the contrast in afternoon temperature between Yarmouth 
and Norwich, Scarborough and Hull, Whitley Bay and Hexham in 

Plate Xla: Cumulus cloud over the Channel, " Armada weather "July forenoon: 

fresh S.W. breeze. 

b: Village cricket at YVombourne, Staffordshire. Early afternoon in fine 
anticyclonic weather, July. 


sunny warm summer weather is particularly noticeable, the sea-breeze 
undercutting from the opposite direction the prevailing warm land wind 
in the manner illustrated above at Seaton Carcw south of Hartlepool. 
On the west coast, gradient wind and sea breeze combine their efforts 
more frequently; the change in the afternoon maxima between Bolton, 
Wigan and Southport, or between Preston and Blackpool, is not 
in general so marked as between Norwich and Yarmouth. Precise 
comparisons are not easily made owing to minor local site factors but 
the following figures will serve to illustrate. 

Approximate Mean Daily Maximum for July: 
West and East Coast Counties. 

Blackpool 657°F. 

Yarmouth 66 -fl°F. 

Southport 66-5 

Lowestoft 66 -9 

Hutton 67-0 (10) 

Norwich* 70-2 (18) 

(Nr. Preston) 

Bolton* 67-9 (25) 

Geldeston-Bungay 70-2 (10) 

(♦adding 1° for correction to 

Cambridge 714 (<jo) 


(♦adding o-3° for correction 10 sea level) 

Distances from sea in miles in brackets. 

It follows however, that if the gradient wind is from the E. or 
N.E. and thus reinforces the sea-breeze from the North Sea the effects 
may be felt much farther inland. The map below ilJustrates the 
distribution of mean daily maxima at places close to sea level, for 
July 1934, a very warm fine sunny month with a considerable number 
of days during which light easterly winds prevailed. In the warm dry 
August of 1947, similar effects along the coast of Northumberland were 
even more marked. Farther north the average afternoon temperature 
at Nairn, Inverness and Fortrose was appreciably lower than at Banff 
or Lossiemouth, evidently because of the tendency in this exceptionally 
warm month in the Highlands for the sea breeze to be drawn in to the 

Plate XI la: Lliwcdd from Crib y Ddysgl, Snowdon. " Autumn anucyclone." 

Afternoon, October 14. 1945. looking south. Lighi E. wind; smoke haze reaching 

8500 ft., Cadcr Idris in distance. 

b: The low stratus of an autumn day in the hilly Outer Hebrides: 
Harris. October, 1947. Remnants of the jetty of abandoned whaling station. 
C.B.S. M 



mnei Moray Firth. Normally Fortrose and Nairn at the observing 
stations tend to be slightly warmer in the afternoon than Banff. 

Fio. 5a 
Mean daily maxima over the British Fslcs, July 1934. Isoplcths al intervals of 
a°F. ; a fine warm month with marked sea breeze effects, especially on cast coast 

If the sea breeze and gradient wind combine, it is occasionally 
possible on warm days for a slight but perceptible effect to reach 
Cambridge, 50 miles inland from the Wash. The freshening wind is 
accompanied by a slight fall of temperature and increase in humidity 
lasting an hour or so. On hot summer afternoons with an easterly- 
breeze a certain freshness about 5 p.m. in London may possibly be 
attributable to the same cause. Similarly the afternoon breeze from 


the Severn estuary may penetrate some distance towards Gloucester; 
but effects such as these are infrequent. 

It has already been shown how, in summer, the decreased con- 
vectional activity and the slight descent of air near our coast on summer 
days result in slightly greater freedom from cloud and a tendency for 
sunshine duration to be, on the whole, about 10% greater on the coast 
than inland. 

Lakes in the British Isles are scarcely large enough to have note- 
worthy effects on wind, although those which lie longitudinally in 
Scotland and N. England resemble the estuaries, forming channels 
along which the wind moves more freely. The winter north-easter 
along Ullswater, and the sharp hail squalls sweeping down Loch Shin 
in Sutherland from the north-west are extremely perceptible if one 
happens to be there at the right time. The slight local air movements 
arising round the shores of lakes and large rivers on calm nights 
however are significant; in the same manner as on the seacoast, they 
provide just sufficient air movement close to the shore to mitigate the 
incidence and severity of frost on clear nights, so long as the waters 
remain unfrozen. But the effect of the freezing over of a lake such as 
Derwcntwater can be suspected by comparing the figures of minima 
on exceptionally cold mornings at Keswick, adjacent to the lake, and 
at Penrith some miles distant. 




8 May 1938 



Lake open 

21 Jan. 1940 

Lake frozen 

26 Jan. 1945 



Lake frozen 

29 Oct. 1946 



Lake open 

30 Oct. 1949 



Lake open 

Some examples of minima on cold mornings at Keswick and 
Penrith, Cumberland. 

Local breezes of another kind arise wherever there are hills; 
especially if the hills arc bare of woody vegetation. On a clear evening 
the loss of heat by radiation from the surface of the earth is particularly 

1 62 


marked around sunset. Radiation from the tops of exposed hills and 
plateaux leads to the cooling of the adjacent layer of the air below that 
at the same level away from the hill-top. Hence the cooler and denser 
air tends to slip downward at once, and as radiation continues and 
more air is cooled, a quite perceptible shallow current of air gravitates 
down the hillsides, gradually undercutting the warmer air below. 
Such winds are known as 'katabatics' by the meteorologist; the West- 
morland farmer at the foot of the Crossfell escarpment knows them as 
'the fell wind' which often sets in as a gentle breeze on a clear quiet 
evening towards sunset. In country of gentle relief such movements 
are slight; shallow currents near the ground can be detected by means 
of cigarette smoke in places such as the chalk downs. Sometimes, for 
example in the Chilterns, the smoke from autumn bonfires can be seen 
creeping down the hillsides until sometime later in the evening the 
valleys are filled with a cooled mass of air from which radiation con- 
tinues; not uncommonly the result as many will at once recall, is a 
valley fog early next morning. 

Katabatic flows in our country of gentle relief rarely attain any 
notable intensity, but they serve to reinforce the land breeze on occasion 
along the Scottish sea lochs, and they may play some part on the North- 
umberland and other coasts. They appear too to play some part on 
quiet nights in regard to the spread of smoke haze from cities such as 
Manchester, Sheffield and Leeds, as we have seen. 

All thermal effects of this kind are reinforced when the ground is 
snow-covered. Freshly fallen snow includes a great deal of air and 
hence, like dry sand, it is a bad conductor of heat from the subsoil. 
Moreover, snow not only forms a good reflector of any radiation 
received during the day; a snow surface is also a very good radiator at 
night. Hence we find that the land breeze is more marked on a clear 
night when the land is snow-covered, and the Tyncside smoke is carried 
far out to sea as the Norwegian skippers know. Descending air-streams 
from the mountains are more vigorous and the consequent ponding of 
cold air is more marked. On a small scale indeed we reproduce in 
Britain all the effects which become really noteworthy along the 
Greenland and other polar coasts. Here however the katabatic 
becomes a roaring gale down the fjords, although its origin is attested 
by the fact that it is often not more than a thousand feet deep and 
fresh-fallen snow remains undisturbed on the higher ledges of the 
coastal mountains. 



The torpor oj (he year when feeble dreams 
Visit the hidden buds, or dreamless sleep 
Holds every future leaf and flower . . 


The tables and maps above (pp. 159-60) have already drawn 
attention to the nature and amount of the temperature difference 
to be expected on warm summer afternoons between seaside and inland 
stations. We appear to be particularly perceptive of small variations 
of temperature and humidity around 6o°-70°, for reasons to be dis- 
cussed in Chapter 15 (p. 291). Small variations around the lower 
forties are important with regard to the progress of vegetation in spring. 
Small variations in the lower thirties arc very significant as they often 
determine whether delicate plants, or crops at critical stages of their 
growth, are to be damaged by frost. At still lower temperatures small 
differences in location may determine whether peach trees for example 
are killed to the roots or not; the lethal temperature for a number of 
familiar cultivated plants lies between 25T. and zero (olive 20°F., 
peach about o°F., macrocarpa about 23°F., for example). Even winter 
wheat may suffer considerably if it is insufficiently protected by snow, 
as in Eastern France early in the winter of 1946-47. 

We may review the character and amount of local variations in 
temperature. These small but significant variations in air temperature, 
from day to day and from place to place, can be largely ascribed to 
those factors which lead to decrease or increase in the freedom of 
movement of the surface air at times when inward or outward radia- 
tion is active. Quiet clear sunny weather with little air movement is 
often associated with the passage of a large anticyclone, and the clear 
skies by day and night result in a wide daily range of temperature. 


Inland position with less air movement than by the sea is also conducive 
to a greater daily range. Valley locations, again leading to restriction 
of air movement give greater daily ranges than neighbouring hill tops. 

Conversely, heavily clouded skies, vigorous air movement in more 
stormy types of weather, exposed hill-top location, arc all conducive 
to decreased daily range. If we add to these any other features allowing 
for free air movement, such as lakes, we shall find that adjacent to 
their shores the daily range of temperature is slightly less than in 
similar locations lacking such opportunities. 

Within extensive woodland, although the movement of air is 
impeded, the result of the impediment to inward and outward radia- 
tion in clear weather is a decreased day-to-night range of air tempera- 
ture compared with the country outside. Small copses and belts of 
trees are however more common in Britain; together with hedges, 
they impede air movement but do little to shade the ground as a whole. 
Hence the irregularly wooded character of the verdant countryside we 
see from any hill-top leads in many instances to increased range in 
some fields while elsewhere some protection is afforded from extremes, 
depending on the degree of shade and other factors, notably minor 
irregularities of the local relief. 

Topographic barriers of every kind from mountains to stone walls 
hinder the free movement of surface air. Further, in regions of irregular 
relief extensive ponding of cold air in all hollows becomes very notice- 
able. On a clear evening the mechanism is evident; from the neigh- 
bouring higher ground there is a shallow gravitational flow of air 
which is often at a maximum a little after sunset, as we have already 
seen. Gradually the warm valley air is undercut; it rises and spreads. 
After the cooling of the uplands has continued for some hours 
sufficiently to maintain the katabatic flow, downward radiation from 
the warmer air which has been spreading at higher levels begins to 
balance the loss of heat from the surface and hence we do not as 
a rule find that the katabatic flow, resulting from the cooling by 
radiation of the uplands, is maintained throughout the night; it 
often stops before midnight, unless the summits are thickly covered 
by snow in which case outward radiation predominates for a longer 

Readers must not assume that these katabatic flows represent 
vigorous breezes. They are often very shallow, only a few feet at most, 
and very gentle; but die aggregate effect is considerable over a large 


area. To quote from a paper by E. L. Hawke with regard to the frost- 
hollow at Rickmansworth "on quiet clear autumn evenings when 
garden bonfires of damp leaves are burning, it is common to see during 
the hour after sunset, rivers of white smoke slowly winding their way 
down into the northern strip of the valley from points in all four 
quadrants of the compass ... the estimated speed of the air flow 
seldom exceeds about 2 m.p.h. It docs not take long for a lake of 
cold air 30-40 feet deep to accumulate." 

Contours in feet 
Built-up areas stippled 

FlO. 53 
Location of Ushaw, Durham and Houghall meteorological stations 

As the "ponds" of cooled air accumulate, they in turn continue to 
lose heat to the clear sky above; within them, again, the coldest air 
sinks to the bottom. By dawn the result is that the valley-floor minima 
are sometimes many degrees below those on the neighbouring hills. 
These differences, as one might expect are enhanced when not only the 
summits but also the valleys are deeply covered by snow. The present 
writer estimates that the difference of temperature between lull and 
valley over 150-200 feet, following a calm clear night with a deep 

1 66 


fresh snow-cover, is likely to be nearly three times as great as it would 
otherwise have been. Examples are given below. 

The extent and character of these effects on minimum temperature 
arising from differences of relief within a broad valley can be very well 
shown in the neighbourhood of Durham. Here the river Wear has 
incised its course sufficiently to make a valley often hall' a mile or more 
in breadth within the larger valley, six or eight miles broad, lying 
between the flanks of the Pennines and the East Durham upland of 
magncsian limestone. Three meteorological stations at different 
levels exist as the map (fig. 53) shows. Over a period of fifteen years 
the following average maxima and minima were obtained: — 











F M 

M J J 

O N D 

42-4 42-6 47 3 50-8 56-5 63 o 66-6 66 5 61 1 53 2 46-8 42-4 
43-1 43-7 48-3 51 -6 57-1 63-7 678 67-2 61 -g 54-7 47-9 431 
44-0 44-7 49-3 52-5 57-9 64-7 68-6 68-1 63-0 55-7 48-8 43-9 

326 329 35 3 37-7 42-3 47-2 5>-8 5 1-1 47-5 4' -8 37-3 33 9 
31-1 32-0 33-7 36-8 41-0 46-4 50-9 50-0 46-0 40-0 35 -6 32-1 

23-2 23-826-3 292 32-4 385 44-5 43-7 3 8- 3 3'-3 288 25-2 
20-9 21-5 241 266 29-2 365 41 -6 40-3 35-6 29-1 25-9 23-5 
16-9 186 203 23-8 26-4 33 3 392 37-3 31-7 26-622-7 19-7 

Average Daily Maxima (A) and Minima (B), and Average 
Extreme Monthly Minima (C) for three adjacent stations, 
Ushaw (594 feet), Durham (336 feet), Houghall (160 feet). 

Period 1925-1940. Mean Miiuma at Durham are omitted as the 
observations refer to different terminal hours (2ih. not ojj.) 
and are therefore not strictly comparable. 

Looking over these figures it is evident that throughout the year 
the average daily range of temperature is lower on the hill-top than 
in the valley, and the effect of relief is more marked with regard to 
the minima. These arc lower in every month than on the hill. But not 
all nights are clear and calm. Many are windy and cloudy, and if on 
a cloudy night with, say, a strong west wind blowing, the minimum 
temperatures are compared it will be found that they conform closely 


to that we should expect from the known expansion-rate of air rising 
and falling in its turbulent course over the surface. The adiabatic 
lapse rate, considered in Chapter 3, is about 5-4° per 1,000 feet; on 
a windy and cloudy night therefore Ushaw, 434 feet above Houghall 
can be expected to be 2-3° cooler, a value borne out by the observations. 

But inasmuch as the valley station is generally 1 -6° cooler than the 
hill-top, and as clear skies predominate on about one night in three, 
it follows that to make up for the windy and cloudy nights which are 
warmer in the valley, the minima on clear nights must average about 
4 cooler in the valley than on the hill. This is closely confirmed by 
the observations, as the average extreme minima show. 

The exceptional effect of outward radiation on clear calm nights' 
with a deep snow-cover impeding the conduction of heat from the 
soil below is shown by the values of the screen minimum temperature 
on occasional winter nights, some of which arc quoted below: — 




















Ushaw, 594 ft. T. 
Durham, 336 ft. 
Houghall, 160 ft. 






















Exactly similar effects with much the same magnitude arc found in 
many other districts. At Malvern, overlooking the Severn valley from 
the flanks of the narrow north-south ridge of the Malvern hills, the 
incidence of frost is much less acute than at Pcrdiswell near Worcester 
or at Droitwich, or for that matter even Cheltenham on the slightly 
higher ground towards the foot of the Cotswolds. 

Pcrdiswell, incidentally is the Worcestershire County Agricultural 
station; it lies two miles from Worcester, about a mile from the river 
Severn and forty feet above it. It is however, important to recognise 
in all these discussions that much depends on the size of the open 
hill-top from which radiation leading to downward flow of cool air 
proceeds. Moreton-in-thc-Marsh (450 ft.) lies higher than the station 
at Malvern, but is in a valley between wide stretches of open though 
cultivated Cotswold plateau; it is accordingly far more liable to frost 
by virtue of its position. Bromyard (393 ft.) is also a valley site 
liable to frost, a fact which appears to have been well known in 



I 926-1 940 





A T. 

33-4 33-7 34'9 3 8> 5 43 -1 483 52'5 5' 6 47"6 41 -6 37-6 34-2 
35-3 35'5 37-4 4°"7 45 "4 5' ■' 55 "4 54 ' 8 5' '» 44 8 4° - 2 3 6 "' 


B °F. 

20-8 224 22 3 26-1 30-9 37-1 41 -7 40-4 33 26-6 23-9 21 -i 
23 9 26-6 268 32 4 36-6 44-3 48-9 478 42 34 4 30-9 249 


Average Daily Minima (A) at Pcrdiswell, near Worcester (94 feet) and 
Malvern (377 feel), and Average Extreme Monthly Minima (B), 1926- 


the seventeenth century as it is mentioned in a treatise on 
Herefordshire orchards written in 1657. 

Now there is a long period of the year, notably in autumn and 
spring during which in Great Britain the arrival of one of the cooler 
air-masses may well result in a clear night with widespread minima 
under 40°. Under these circumstances the local geographical factors 
play an extremely important part. For example if at Ushaw and 
Malvern the minimum falls to 38 there is a considerable chance that 
at Perdiswell or Houghall the air temperature will fall to 32 ° with 
consequent damage to many plants at a critical stage. Early in May 
the Worcestershire fruit trees arc in blossom; later in the month those 
at Durham, together with the newly-sown potatoes, are vulnerable. 
Over a period of years the average length of the season free from frost 
(minimum air temperature below 32 ) is about eight weeks longer 
at Malvern than at Pcrdiswell; and there is evidence that the 
particular location chosen for the Droitwich climatological station is 
slightly more liable to frost than Perdiswell, five miles to the south- 
westward. These general features of the Severn lowland are supported 
by other stations such as Prestwood, Stonehouse and Defford with the 
Cotswold upland airfield at Little Rissington for comparison. 

All over the country these effects have been recognised by the 
intelligent farmer long ago. Kentish orchards thrive better on the 
slopes; around Cambridge a local company makes extensive use of 
the slightly rising ground near Haslingfield. In Worcestershire the 
belt of orchards girdling Brcdon Hill may be mentioned. In Scotland 


the slopes of the middle Clyde valley are more favoured dian the 
frosty upland basins of Peebles-shire. In many other districts however 
the favourable sites for cultivation of fruit can be recognised at once by 
anyone with a little experience of the manner of incidence of frost; 
coupled with a recognition that it is equally necessary to avoid the 
higher exposed and windy summits. 

It has already been 
pointed out that among the 
factors discouraging the 
ponding of air is the prox- 
imity of lakes, estuaries or 
the sea. Sloping ground in 
such positions is often partic- 
ularly free from frost unless 
there is considerable oppor- 
tunity for drainage of air 
from uplands in the interior. 
Pendennis Head, small in 
area and surrounded by 
the sea on three sides, has 
fewer frosts than Falmouth 
Observatory a mile away 
although die latter lies 167 
feet above the sea and on 
a favourable slope for air 
drainage. Around the 
Observatory Falmouth in 
turn is slightly less liable to 
frost than Fowey, and both 
are much freer from frost 
than Bude, where as any 
map will show the air can 
drain downward from quite 
a wide area. On the Devon- 
shire coast, the Sidmouth 
station, at the mouth of a 

narrow wooded valley, shows slightly less tendency for extremes of 
frost than that at Seaton, lying at the mouth of a broader, longer 
and more open valley with a nearly flat floor. 


1 (/ ftv ^5s, 

) BREeKtAND \ \~\ 


7=n 4&A ///> 

c L8' 

ZuJ y*C3^~ 20 

Mid -East Kent: _ 
Below zero inland 

Fie. 54 
Approximate distribution of minimum 
temperatures in degrees F. for East 
Anglia on ao January 1940, with iso- 
pleths sketched at 4 intervals. Inland 
minima in the Weald of Kent fell con- 
siderably lower owing to the presence of 
a deep snow-cover, although at Biggin 
Hill (567 ft.) the minimum was 12 , of 
the same order as that above 500 ft. on 
the Chiltcrns. In the Rickmansworth 
frost- hoi low zero was recorded 



Almost all over Britain the local relief is appreciable. Officers of the 
Forestry Commission have noted some remarkable frost-hollows among 
the undulations of the central Weald of Sussex. Elsewhere severe frost 
occurs in the boggy hollows among the Eden valley drumlins of West- 
morland. Moreover, quite small barriers — even walls — may serve either 
to divert or to impound the gentle downward streaming of the cool air 
on open slopes at night. Those who wish for early skating would do 
well to recall the simple prescription by one of our most eminent meteor- 
ologists, Sir Gilbert Walker, at Simla many years ago. He recommended 
that an earth bank should be erected on the border of some tennis courts 
lying below a certain shady slope, thus damming up the flow of cool 
air and successfully retaining it over a very shallow flooded surface. 

It will Uius be evident that the incidence of frost is extremely 
variable, depending on distance from the sea, degree of cloudiness and 
local air movement, relief and several other factors, notably the 
nature of the soil and vegetation. 

Before we consider these further points the Figure (54) above showing 
the incidence of extreme minima based on the official records through- 
out Eastern England on the very cold morning of 20 January 1940 
may serve to illustrate as far as the limited observations permit the 
degree of variation that can be expected on an occasion when there 
was, north of the Thames at most a very thin powdering of snow and, 
as far as can be judged, an almost completely calm cloudless night. 
Note how adjacent to the coast the minima were very much the same 
throughout, whereas inland, north of the Thames, they ranged down 
to 20 colder in exceptional hollows, notably that at Rickmansworth. 

On the same night however East Kent and S.E. Sussex lay under a 
deep snow-cover. Although the coastal minima at Dover, Margate 
and Hastings were very similar to those in Norfolk, inland minima 
were very much lower. In hollows inland, they ranged down to -6°, 
for example at Bodiam in Sussex, with -4 at Canterbury. It is evident 
that when the Downs arc snow-covered very low minima indeed can 
occur in such locations; in January 1947, -6° was again recorded, 
at Elmstone in E. Kent. In East Yorkshire, very low minima are 
recorded from time to time at the foot of the Wolds. This helps to 
explain why the minima reported from Bridlington in winter are much 
lower than those at Scarborough. 

For the morning of 20 January 1940 the isoplcths may be con- 
jectured to have run somewhat as shown. It is however rare to find a 


night on which die prevailing condidons over an extensive area are 
sufficiently uniform for relief and other factors to play their full part, 
but with the aid of the data in the Monthly Weather Report many will 
find it interesting to attempt the construction of similar maps for such 
extreme nights as 3 March 1947. 

This map will serve to draw attention to other factors governing the 
incidence of extremely low minima. The first of these is the character 
of the soil. The surface of the earth loses heat by radiadon at night, 
but this is to some extent offset by the conducdon of heat to the surface 
from below. The surface soil layer, i.e. the top two or three feet, 
varies considerably in conductivity. If the soil is coarse-grained and 
includes a good deal of air, it is a poor conductor. Hence sands and 
sandy soils by virtue of their low conductivity and rapid drainage give 
surface temperatures on clear nights well below those obtained on more 
normal clayey loams which not only contain less air but are better 
conductors by reason of their longer retention of water. Incidentally 
a deep fresh snow-cover contains a great deal of air and is a very poor 
conductor indeed of warmth from the ground below. Hence as we 
have seen exceptionally low minima can be expected when the sky 
clears following a heavy snowfall. (Table p. 167). 

In the map above it will be seen that the lowest minimum in East 
Anglia occurred on the Breckland near Thetford. This is an area of 
particularly light, well drained sandy soil and on clear nights minima 
upon it are generally about six degrees lower than on gravelly loam 
at Cambridge, unless the sand has just been wetted by heavy rain or 
else there is a continuous snow-cover. The same factors operate 
elsewhere; at South Farnborough minima tend to be rather lower than 
at other inland Hampshire stations because of the situation in a broad 
valley with a rather light sandy soil. 

! 94 1 May 
















Cambridge (Univ. 







The effect of sandy soil at Lynford on the Norfolk Breckland. 

Night Minima at Lynford and Cambridge with a Scandinavian 

anticyclone following dry weather, May, 1941. 



'933 -'94' 




2-2 2-5 2-6 2'7 3'I 3"I 3"I 37 2-8 2'4 I-g l"9 



6-o 6-i 5-8 6-8 6-i 7-5 6-6 7-6 6-i 5-9 4-7 5-4 


Departures (°F.) of the Average Daily Minima (A) and Average 
Extreme Monthly Minima (B) at Lynford (99 feet) below 
Uiose at Cambridge University Farm (78 feet), 1933-1941. 

Relief and dry, well drained soil combine to provide in a small valley 
in the Chilterns near Rickmansworth one of the most exceptionally 
frosty locations yet known in the British Isles. The extreme minima 
apply, it should be noted, within a shallow layer of air on the floor of a 
dry valley in the chalk, surrounded in the main by bare grassy slopes 
rising 100 to 150 feet above. The area concerned is very small, about 
half a mile long and perhaps a hundred yards wide; but there is no 
doubt that similar exceptionally cold pools occur here and there in 
neighbouring valleys. Phenomenally low temperatures at all times of 
year have been recorded at Rickmansworth ; quite commonly on clear 
nights minima are io°-i2° below those of nearby London suburbs. 
Our knowledge of the possibilities of British frost hollows owes a great 
deal to the enterprise of a long-standing Honorary Secretary of the 
Royal Meteorological Society, Mr. E. L. Hawke, in establishing and 
maintaining a record on the floor of this remarkable valley, in which 
dahlias may be lost in mid-August whereas a quarter of a mile away 
they survive until November. 

In such a constricted valley temperatures by day tend to rise 
higher than elsewhere; and on a calm clear and dry day, 29 August 
1936, the exceptional daily range of 50-9° was observed, with a maxi- 
mum 85 , minimum 34° within ten hours. A daily range of this 
magnitude is almost worthy of the Sahara. 

Still another factor in the incidence of frost, considered as air 
temperature below 32 , is the nature of the vegetation cover. Studies 
at a number of stations have shown that the lowest minimum tem- 
peratures tend to occur over long coarse grass. Bare ploughed earth, 
many will be surprised to hear, does not in general give such low 
minima. By and large it would appear that centuries of drainage and 
cultivation have on the whole tended to improve the climate. The 
occurrence of low minima over long coarse grass is probably because 


Rickmansworth frost-hollow 
1936 Auausb 29 

the grass-blades themselves 
radiate heat, and their cool 
surfaces chill the air be- 
tween the grass blades 
while impeding its move- 
ment; also the grass stems 
are not good conductors 
of warmth from the roots 
below. In many districts 
the ill-drained boggy hol- 
lows are often the first to 
experience the effects of 
frost in autumn, and are 
very slow to warm up in 

From what has been 
said it will be evident that 
local variation in minimum 
temperature on severely 
cold mornings can be very 
large; and that really low 
temperatures often apper- 
tain merely to small 
patches of ground. Hence 
much discrimination must 
be used in estimating the 
liability to frost at any 
particular place in 
comparison with nearby 
meteorological stations. 

Great cities, as the map (p. 169) reminds us, also tend to have higher 
minima than the open country. Two factors arc at work; first, the 
surface winds are lighter owing to the frictional drag imposed by the 
buildings. Secondly, radiation is absorbed especially in the warmer 
months by the roofs and sides of buildings during the day, and is given 
out at night; in courtyards, streets and enclosed spaces the result is 
that temperature falls much more slowly during the evening, and in 
any case does not attain the minima found outside the built-up area. 
Nothing is more familiar to the young London motorist in a hot 

Fie. 55 

Thermograph trace showing one of the 
greatest daily ranges on record in England; 
minimum 34 - o°F., maximum 84-gT., range 
50-g°F. (after E. L. Hawke, Q.. J. Roy. Met. S., 


summer than the evening return across the cabbage-fields of Middlesex 
when it becomes a question whether he and his passengers should don 
their jackets. Coming up the Great West Road however the air along 
the suburban roads at 10 p.m. is still several degrees warmer than over 
the fields, and in Mayfair every door and window stands open in the 
hope that some cooler air will arrive. 

It follows that daytime maxima in summer within a city, even in 
parks of some size, are also in general slightly higher than in the 
country; given that the wind is light and the sun powerful. But if the 
streets and courtyards are sufficiently narrow to admit little sunshine 
local patches of cool air remain within them. Long ago it was observed 
that the average maxima at Old Street in the City tended to be slightly 
lower than those in the parks during summer. Compare for the hot 
July 1901 mean daily maxima at Old Street, 74-3 with Regent's Park, 
758; for July 1900 76-8 and 78-1; and for the hot August of 1899 
75-3 and 763. In the same months the mean minima were 58-2, 
55-8; 59-9, 56-8; and 59-3, 57-0 respectively. At Manchester on the 
other hand the Oldham Road record was kept in an enclosed yard of 
some size (60X40 yards) and surrounding buildings are not high; 
invariably in sunny weather higher maximum temperatures were 
recorded than for example in Whitworth Park (16 acres) and these 
in turn are appreciably higher than at stations outside the city 

In winter radiation received by buildings is less, but a certain 
amount of warmth escapes from the heated interior. Outward radia- 
tion is also to some extent hindered by smoke, and slight air movements 
between streets and parks all tend to check the ponding and cooling 
of surface air by comparison with the country. Hence minima are 
again considerably higher in cities, especially among the buildings, 
and appreciable variations develop between one part of the city and 
another. For example in London if there is an overall slight drift from 
the north, the minima in the northern suburbs approximate more 
closely to those in the country than those in the southern suburbs. 

In our excessively urbanised country the great cities undoubtedly 
form a very prominent element of the British scene and their climatic 

Plate 25 

Rowsley, Derbyshire: heather moor, September. Fine-weather cumulus in early 

autumn, cloud base about 3,000 feet; cloud tending to decrease in amount eastward. 



B. A. Owh 


effects are coasiderable. The amelioration as far as frost is concerned 
in London is now quite appreciable. A minor effect has also become 
noteworthy in the last 30-40 years; fogs in the centre of our great cities 
are not quite so great an impediment to traffic, at street level, as they 
formerly were. In bad London fog visibility outside Bedford College 
in Regent's Park and along Kensington Gore may decrease to 10 yards; 
but along Oxford Street one can see 100 yards or more. There is little 
doubt that this effect arises from the very slight warming of the air in 
the streets, just sufficiently to absorb a part of the suspended moisture. 
The reputation for fogginess of the great suburban by-passes owes much 
to the fact that they often run through the less-favoured damp clays 
and secondly to their freedom from buildings. 

The fact that the nocturnal fall of temperature is checked by the 
radiation from walls and buildings has long ago been recognised. The 
Elizabethans may have been among the first to observe the advantage 
when they tried their new flowers and vegetables under the warm 
brick walls of their houses, which in more ways than one gave scope 
for individual initiative. Walled gardens became the fashion; grapes, 
peaches and apricots were ripened on south-facing walls. Even in 
Tcesdale there is a Jacobean walled garden where peaches are still 
grown with success. Eleven hundred feet above the sea at Garrigill in 
the Cumberland Pcnnincs near Alston two plum trees under a south 
wall ripen frequenUy, and provide a rare example of the possibilities 
attached to the provision of some shelter in that otherwise cold district. 
Walled gardens must however be properly designed in order to prevent 
the accumulation of cold air within them at night (here Sir D. Brunt's 
papers should be consulted by those interested). 

In all this we observe one of the most impressive features of the 
British climate. The range of temperature is such that small changes 
of location, aspect and shelter play a very appreciable part in the 
success or failure of fruit cultivation and gardening operations of all 
kinds. The gain resulting from a little extra care in choice of site, or 
in provision of protection from wind or frost, is considerable; in these 
and other respects intelligent forethought has for fifteen hundred years 

Plate 26 

From Red Pike, Buttcrmere looking N.F.. towards Skiddaw, Cumberland: sun and 

shadow, early summer afternoon, June. Straio-cumulus rolls at 3,500 feet very 

characteristic of westerly weather in June; clear above 

CB.S. N 


or more received its reward. Even the several fields of a normal farm 
have each their varying characteristics arising not merely from soil 
but also from the significant local variations in the incidence of frost. 
These in turn make for forethought in regard to type of seed, date ol 
sowing, and even more notable, the characteristics of local breeds of 
stock. It is difficult to resist the conclusion that while co-operation in 
times of calamity is desirable, individual enterprise and the sharpening 
of intelligence could not help but be developed in an environment 
offering such local differences leading to abundant opportunities for 
freedom of choice on every hand. 

Anyone who travels about Great Britain and contemplates the 
varied degree of development of almost any country activity must bear 
in mind the climatic environment and the extent to which under 
differing economic circumstances any given type of cultivation or 
craftsmanship has been in demand. Given care, a very great variety 
of possibilities exists. Grain has been grown in war-time at 1,400 feet. 
Many old orchards in Cumberland remind us that before the days of 
Covcnt Garden's penetration to the nordicrn counties quite a fair 
amount of fruit was grown, and might be grown again. 


Chapters 8 and 9 

Brunt, D. (1945). Some Factors in Microclimatology. 0_. J. Roy. 

Met. S. 71: 1-10. 

(1947). Where to Live. J. Roy. Inst. 10: 1-13. 
Darling, F. Fraser (1947). Natural History in the Highlands and Islands. 

London, Collins' New Naturalist. 
Defoe, D. (cd. Cole, 1927). Tour through Great Britain: vol. I: 71-2. 

London, Davies. 
Department of Scientific and Industrial Research (1948). Atmospheric 

Pollution in Leicester. London, H.M.S.O. 
Gold, E. (1936). Wind in Britain. Q,. J. Roy. Met. S. 6s: 167-205. 
Hawke, E. L. (1933). Extreme diurnal ranges of temperature in the 

British Isles. Q.. J. Roy. Mel. S. 59: 261-65. 

(1944). Thermal characteristics of a Hertfordshire Frost Hollow. 

Q,. j. Roy. Met. S. 70: 23-48. 
LAMB, H. H. (1938). Industrial Smoke Drift and Weather. Q_. J. Roy. 

Met. S. 64: 639-43. 
Lewis, W. V. (1938). Evolution of Shoreline Curves. Proc. Geol. Ass. 49: 


landscape features and their effect on weather (2) 177 

Manley, G. (1944). Topographical Features and die Climate of Britain. 

Geogr. J. 103: 241-63. 

(1945). The Helm Wind of Crossfell. Q_. J. Roy. Met. S. 71: 197-219. 
Minnaert, M. (1940). Light and Colour in the Open Air. London, G. Bell & 

Paton, James (1948). The Optical Properties of the Atmosphere. 

Weather, 3: 243-49. 
Raistrick, A. (1943). The Pennine Walls. Dalesman, V: 5-13, 21-28, 

Salisbury, E. J. (1939). Ecological Aspects of Meteorology. Q, J. Roy. 

Met. S. 65: 337-57. 
Scorer, R. S. (1949). Theory of waves in the Ice of mountains. 0_.J. Roy. 

Met. S. 75: 41-56. 
Spence, M. T. (1936). Temperature Changes over Short Distances in 

the Edinburgh District. (£. J. Roy. Met. S. 62: 25-31. 
White, Gilbert (1931). Journals of Gilbert White. London, Routiedge, 

cd. Walter Johnson, 24, note for 23 February 1770. 




Tour snows and streams 
Ungovernable and your terrifying winds 
that howl so dismally for him who treads 
Companionless your awful solitude 

Everywhere in our highlands we are reminded that above a 
I relatively low altitude the potential yield of the land steadily 
decreases. The conditions of existence of a farming population become 
increasingly precarious, until at length a level is reached at which 
farming of any kind is no longer practicable. 

Over the world as a whole the limit of permanent settlement for an 
inland population is, in general, the cold trceline. Broadly it is found 
that trees in the normal sense of the term do not grow unless the mean 
temperature for at least two months exceeds 50°F. Even then, shelter 
and in higher latitudes reasonable freedom of drainage are desirable. 
In Iceland where the mean temperature at Reykjavik is 52° in July 
and 50 in August an occasional tree can be found on the coastal 
lowlands under the lee of a farmhouse; in sheltered valleys farther 
inland the average height of birches in woods may exceed twenty feet. 
Similar results arc found on the better-drained uplands in Northern 
England and Scotland. Towards the coast however, winds are stronger 
and it becomes very difficult to establish trees at all on exposed uplands 
even at 1,000 feet where the mean July temperature ranges (in Scot- 
land) from 55 c to 52 . It is also necessary to recall the deleterious 
c fleets of salt spray near the coasts, which after severe gales have 
sometimes been observed fifty miles inland. In June, 1938, this was 
noticeable in Hampshire. In Wales we may compare the bare uplands 
of Cardiganshire with the wooded upper Towy valley. Inland, given 
some degree of shelter, occasional trees are found all over Britain 


above 1,500 feet and in the extreme instances may grow up to, or very 
slightly above, 2,000 feet. In the Eastern Highlands of Scotland many 
who know Rothiemurchus forest will be familiar with the last struggling 
pines and occasional birches found at this level on the way towards 
Cairngorm or the Lairig Ghru from Avicmore. In Cumberland a 
small mountain ash can be espied at this level in a gully above Ulls- 
water; its survival owes much to the protection its position gives from 
the nibbling sheep. For similar reasons one or two trees nearly reach 
the 2,000 foot contour in Grainy Ghyll, above Seathwaitc. Farther 
to the eastward in the high northern Pennines there is a very good 
example indeed of the struggles of the trees to survive. Above Ashgill, 
south of Alston, a plantation was established over a century ago and at 
1,600 feet Scots fir can be found forty feet high. Extending up the hill- 
side however the trees rapidly dwindle in size. At 2,000 feet those which 
still survive are rarely six feet high, despite some slight shelter from an 
adjacent low stone wall. Moreover quite a number of the adjacent trees 
are dead, and others arc dying. A compact summary for the meteorologist 
of the effect of wind on vegetation has been given by Sir E.J. Salisbury. 

J,F ,M ,A,M , J, J, A, S, 0,N,D 


- fir<* - Oundle, 147 ft 


J,F ,M,A,M, J, J, A, S,0 ,N,D 

Fio. 56 

'Hie trend of monthly mean temperature at selected stations (1906-T935). 

The average daily maxima and minima for January and July arc shown; 

the average length of the 'growing season' above 42 will be noted 

In Northern Britain the rate of fall of mean temperature with 
height is about i° for 270 feet, based on existing mountain records. 
Accordingly wc can see that the level at which mean temperature 
exceeds 50 for two months is about 2,200 feet under present conditions, 
and hence that although a great part of our uplands are bare, trees if 
planted and protected might be found up to this height above the sea. 



Further, in relatively recent centuries scattered trees were prob- 
ably found in many of our upland valleys much more extensively than 
now. The decrease owes much to the combined efforts of men and 
sheep. One of the first essentials in re-afforestation is a sound wire 
fence, whose rectangular assertiveness is now too familiar in Ennerdalc, 
at Dalwhinnie, or flanking the Plynlimon road as Midland motorists 
bound for Aberystwyth can see. 

Under British climatic conditions we have in any case a remarkably 
low trcelinc compared with many other temperate countries. On 
account of the warmer summer inland we find that in Central Norway 
birch scrub occurs here and diere above 3,500 feet; and diat the pine 
ascends to over 2,500 feet. No recollection is more vivid in the writer's 
mind than that of a well-known Swiss professor of geology who, con- 
fronted at 1,680 feet— the level of Berne— with a wide stretch of the 
Pcnnines between Tecsdale and Weardale, surprisingly declared "this 
is the tundra". Yet our climate permits, within a very few miles, the sur- 
vival of delicate evergreens. We may proceed to illustrate why, within 
the relatively small range of mean temperatures arising from a mere 
two thousand feet of altitude, very marked changes in the natural 
and cultivated vegetation and hence of the scenery are developed. 

Altitude, Temperature and the Growing Season 

In the first place diere is the question of length of the growing 
season. Taking the familiar grasses and other vegetation of western 
Europe as a whole it has long ago been observed that growth begins 
and continues whenever the mean temperature exceeds a figure 
approximating to 42 °F. During the cooler winter months the average 
daily range of temperature under British conditions lies between 
8° and 14 ; and it tends to be smaller nearer the coasts, greater inland. 
Therefore, if the mean temperature is 42 it is broadly true that on 
most nights the minimum will remain above the freezing point. 

At a representative Midland starion such as Oxford the mean 
temperature for March is rather above 42 , which implies that, by and 
large, the grass begins to grow a little before the middle of the month. 
Much depends on whether the month is mild or severe. In the English 
Midlands some no doubt who take note of such things will remember 
mowing the lawn before the end of a mild March such as that of 1938 
or 1945. Others will contrast the very slow oncoming of that inevitable 

mountains and moorlands: effect of altitude 181 

suburban spring-song in the severely cold April of 191 7. In some years 
dry clear afternoon warmth in March is accompanied by a much 
greater dailyrangc of temperature than usual and hence by frosty nights, 
as in 1929 ; yet the mean remains low and grass makes little progress. 

As a whole however, we can take out an approximate average date 
for the beginning and end of the 'growing season' from the curve 
showing the march of average temperatures during the year. At 
normal Midland stations the season of growth can be expected to end 
sometime in November (fig. 56, p. 179). 

It will at once be evident however that somewhere to the south- 
westward, in lowland Devon and Cornwall for example, the mean 
temperature diroughoui the year does not fall below 42 °. Hence even 
in mid-winter growth although very slow does not entirely cease, 
save for the few days of dry, cold weather that normally occur two or 
three times between December and February. Cornish coastal gar- 
deners mow their lawns the whole year round; pastures in general 
remain green and the cattle need but little additional feeding, even 
by comparison with the West Midlands. Narrower belts, almost 
equally favourable, can be found in Pembrokeshire, and even in 
south-west Galloway (cf. Scot. Geogr. Mag. 1946); and in Southern 
Ireland a relatively large area offers similar opportunities. 

But we may at once observe that the curve of the annual march of 
temperature is rather flat, that is, the rise in the mean temperature in 
spring is slow compared with more continental countries. With increased 
altitude the rate of decrease in the length of our growing seasons is 
accordingly very rapid. At Nenthead in Cumberland, one of the highest 
villages ( 1 ,500 feet) in die whole of Britain, enclosed pastures run up the 
adjacent hillside to 1,800 feet. Compared with the lowlands the mean 
temperature is lowered sufficiently to decrease the "growing" season (18 
April-23 October at 1,500 feet) by about ten weeks, and at the limit of 
settlement (about 1,900 feet) the decrease is to 4 May-16 October. 
Hence, whereas on the south-wesi coast of Cumberland sheltered 
corners can be found in which the grass almost remains green through- 
out a normal year, forty miles away it is necessary to feed cattle, if any 
are kept, for six months at the very least; hay is also given to the sheep. 

Lambing time is closely associated with the date when the fresh 
growth can be expected. At Moor House (1,840 feet) the lambs are 
expected about the first week in May, whereas in the lower Yorkshire 
dales early March is customary. Expressed otherwise; in some districts, 

1 82 


a change oflcvcl of 1,800 feet approximately halves the growing season. 
Climatic data from Malham Tarn Field Centre are noteworthy (p. 196). 

The rapid change in the length of the growing season means that 
with a small change of altitude great contrasts are found in but a short 
distance. In 1947 lilac and laburnum were out together on 28 June 
beside one of die highest cottages (1,550 feet) in upper Weardale, 
five weeks later than in the nearby lowlands of Cumberland. It is 
particularly in springtime that the contrasts are most marked for, as we 
shall see, the fall in the average temperature with height is more 
marked in May than at any other time of the year. 

Another effect of altitude becomes very evident from the curves. 
Within the range of temperatures and rainfall characteristic of the 
temperate zone it is broadly true that the intensity of growth depends 
on the amount by which the mean temperature of the warmest month 
exceeds that (42 ) at which growth may be said to begin. With a July 
mean of 70 the rapidity of growing and maturing of crops is perhaps 
twice that which we can expect with a July mean of 56 . These figures 
are however not to be regarded as exact; the amount and intensity 
of sunlight, the frequency and character of rainfall and the retcntive- 
ness of the soil ail play some part. It is evident however, that a rough 
relationship exists between the rate of growth and die consequent 
yield, and the area lying between the curve of the annual march of 
mean temperature and the line representing 42 . This explains at 
once why to our eyes the progress of vegetation is noticeably rapid in 
more continental climates such as that of Central Europe, despite the 
greater severity of winter frost. Further, in Britain a cold late spring 
is for many farmers a greater calamity than a severe winter. 

Now a little thought will show diat where the curve of the annual 
march of temperature is rather flat, for example in Scotland, a small 
change of altitude makes a much greater difference to the area 
above the 42 line (Fig. 55) than would result from a similar change in 
a climate such as that of the Eastern United States or Vienna. Hence 
we find that under British climatic conditions not only is there a sharp 
decrease with height in the length of the growing season; the decrease 
in the rate of growth and the yield of crops is again rapid. In the 
Northern Pennines for example the average yield per acre for oats at 
1,000 feet may be put at scarcely half of that near the coast. 

Further; the effect of seasonal variations is much more marked on 
higher ground. If in a cool summer the overall average temperature 


of the three summer months is 58 instead ot 60 ", the yield of grain, for 
example, can be expected to decrease. But a thousand feet higher 
up the mean (other factors disregarded) will be 54 instead of 56 , and 
without doubt a decrease of 2 from 56 will tend to be associated with 
a greater proportionate fall in yield than a similar decrease from 6o c . 
Moreover, in a cool and cloudy summer with excess of rain, the amount of 
that excess is likely to be greater on high ground; hence quite apart from 
seasonal variations of temperature, die year to year variadons of rain- 
fall also render agricultural operations very much more precarious on 
higher ground. Thirdly, in a cool cloudy summer it is more often than 
not found that the aggregate deficit in mean temperature is greater on 
high ground than near sea level. Data for several summer months which 
were persistently cool and cloudy throughout are given below: — 

Cool and Cloudy Summer Months 

Departures of the Mean Monthly Temperature below normal at high and 

low level stations 


July 1920 August 1912 September 1918 

Princctown 534 : -3-1 51-2 : -5-3 50-8 : -29 

(1359 ft-) °F- 
Teignmouth 60-2 : -1-9 57-4 : -47 57-0 : -17 


July 1920 July 1922 August 1922 

Brat-mar 50-6 : -4-5 50-6 : -4-5 507 : -3-0 
(1114 ft.) 

Perth 56- 1 : -30 556 : -3-5 556 : -2-1 
(76 ft.) 


July 1936 August 1943 August 1946 

Moorhouse 515 : -1-3 517 : -o-6 49-6 : -2-7 
(1840 ft.) 

Penrith 575 : -06 57-3 : +0-4 55-3 : -1 -6 
(559 ft-) 

Given longer records we could with a little trouble almost express in 
figures the risk attached to arable farming on high ground. Shrewd men 
however, have long ago expressed the risk in the rent. To a certain 
degree the upland farmer's disappointments resulting from cool and 
cloudy summers are onset by the gain if the summer happens to be dry 
and fine. In warm dry anticyclonic weather with little wind the mean 


temperature on the high ground approaches much more closely than one 
would expect to that at low levels. This is partly because of the nocturnal 
tendency for the coldest air to seek the lowest level. Two or three weeks of 
warm dry weather at the right time arc proportionately even more bene- 
ficial in the uplands than below. The effect is well shown by the follow- 
ing figures from upland and lowland stations, covering months in which 
warm anticyclonic weather prevailed for the greater part of the time. 


June, 1940 

(1840 ft.) °F. 
Mean max. and 

56-1 : -H 6-7 
67-0, 45-2 


(559 ft.) °F. 
Mean max. and 

59-4 i +47 
71-8, 46-9 



August, 19.17 

('359 ft-) 

637 : + 6-8 
7'-5. 55-8 

(202 ft.) 

66-7 : + 5-6 
78-6, 548 

At other times of the year the incidence of calm quiet anticyclonic 
weather also results in relatively greater benefit to the uplands. This is 
shown by data for March 1933 and 1953 and December 1935 — when 
anticyclonic weather prevailed for two or three weeks. 


Mar. 1933 

Dec. 1935 

Mar. 1953 

Moorhousc (1,840 ft.) 

Monthly means and departure 
from normal 
Mean max. and min. °F. 

457. 327 





479. 29-2 

Penrith (559 ft.) 

Monthly means and departure 
from normal 
Mean max. and min. °F. 

5 1'5. 35-3 



382, 286 




The converse effect is seen when the weather is prevailingly windy 
and air derived from polar sources with a long sea travel predominates. 
The lapse-rate is thus rather high in the surface layers. Under such 
conditions, as we saw in an earlier chapter, everywhere on our 
uplands and mountains cloudiness is greater and showers are more 
frequent while at the same time surface winds are stronger. Hence 


the difierence in temperature with height is more marked than usual, 
especially in spring when the sea surface is cool and polar air reaches 
us with less modification than would otherwise be found. 


M. Max 

M. Min. 


Moorhousc (1840 ft.) 
Normal May, °F. 
Dull, wcttish cool 
May, 1932: (2-4° below 


5' "8 



Penrith (559 ft.) 

Normal May, °F 

Dull and rather cool May, 

1932: (1-7° below normal) 




67, 26 

Note that the average daily maximum on the upland fell below 
the normal to a greater extent than in the lowland. 

That such differences are not confined to the Pennines is quite 
evident from Dartmoor, or even the Chiltcrns: 


Princetown (Dart- 
moor) 1 359 ft. °F. 

3 Lowland stations 

March 1945 

Centrally warm 

and dry 

May 1946 
Generally cool, wet 
with Js'.E. winds. 

January 1940 

Exceedingly cold; 

often quiet and 


above normal 

2-3 above 

2-4 below 

1 • 1 below 

6-3 below 

8-6 below 


Whipsnade 720 ft. 
(no normal 
available) °F. 

Cambridge, 41 ft. 

March 1945 

May 1946 

February 1947 

V. cold, dull, 

E. wind. 


46-5, 4-2 
above normal 


5' -8, 17 
below normal 


28 5, 11 -2 
below normal 

Whipsnade — 

1-8 Warm 
dry month 

3-3 Cool 
weltish month 

2 -6 Cold and 
very cloudy 
month, occa- 
sionally clear 
and calm. 


From all that has been said it will be evident that the effect of 
increased windiness and freedom of air movement in our uplands is 
more pronounced with regard to the lowering of day-time maxima 
than as regards night minima. All walkers know this. 

Wc may now generalise. First, the aldtudinal rate of change in 
the potentialities of our island climatic environment is especially 
rapid; as a result we have an unexpectedly low climatic tree-line 
bearing in mind the fact that delicate evergreens can flourish in the 
open at many coastal resorts. The average yield of crops falls oft* 
rapidly with height, and the variability of the yield increases. Hence 
for all arable crops a level is soon reached at which the risk of wasted 
labour in cultivation is too great unless it is offset by high prices, for 
example in war-time. 

The rate of change of climatic environment is so rapid as to call 
for a marked degree of adaptation in local breeds of cattle, horses, 
sheep and other domestic animals. These all form an essential part of 
the make-up of our British scene. The Eastern Aberdeenshire up- 
lands are not complete without the presence of an extremely resistant 
and hardy terrier whose hairy coat defies both wet and cold. The 
Lakeland Herdwick sheep are adapted like no other breed to the 
peculiarly variable winter with its intermittent spells of excessive rain. 
Among sheep, the Lonk of the Lancashire fcllsidc will stand rain, but 
is not so adaptable to heavy snowfalls as the Swalcdale — a particu- 
larly interesting feature bearing in mind those north-eastern districts 
in which our heaviest snowfalls generally occur. Adaptations of 
farming technique are called for on every hand; and those farmers 
succeed best who in the long run combine sharpness of wit and 
readiness to adapt their actions to the weather with the needful 
caution before committing themselves and their dependents wholly 
to any given policy. 

Altitude and the Incidence of Frost 

In the previous chapter we discussed the effects of relief, of soil 
and of proximity to water-bodies with regard to the local incidence and 
severity of frost. Hill slopes and lower summits are freer from frost 
than hollows. As we ascend however we generally approach the region 
in which the whole air-mass is frequently below the freezing-point. 
To offset the diminished frequency of frost arising from upland position 


on those quiet clear nights when the cold air slides downhill and local 
inversions develop, wc have an increased frequency of days when even 
with the wind blowing vigorously the temperature remains below the 
freezing-point. Consider for instance an overcast day of east wind 
such as 29 January 1939; this was a day of helm wind along the 
N. Pennincs (p. 149). 

°F. Maximum 


Dun Fell (2735 ft.) 


23 5 

Moorhouse (1840 ft.) 



Durham (336 ft.) 



Manchester (125 ft., urban) 40 


On balance the greatest freedom from the effects of frost at inland 
stations is found on slopes at intermediate levels between perhaps 
200 and 500 feet above the sea. (CI. die data for Ushaw and 
Malvern on pp. 166, 168). Higher up, there comes a point, roughly at the 
2,000 feet level in N. England where the mean temperature of the 
coldest months can be expected to fall below 32 more often than not. 
Here is the region of scurrying low cloud and pitiless raw wind, as 
those who frequent our high moors in winter know. The impacted 
supercooled droplets from the low stratus cloud build up frost-feathers 
on the windward side offence-posts and wire, walls and cairns; snow, 
useless for ski-ing covers the ground more often than not. The uniquely 
exhilarating horribleness of such weather is the more appreciated 
when one knows that two hours distant one finds the ham and eggs. 
But it becomes grimly testing when as in 1947 the persistent wind, 
snow and lack of food led to complete exhaustion of so many sheep, 
who could not rally when die thaw came. 

At intermediate levels however upland valleys and basins can also 
be found offering with greater frequency than elsewhere the possibility 
of air drainage from snow-covered uplands. Probably these hollows 
arc die site of the most severe extremes of temperature wc experience. 
In the Highlands diere is still much room for determining which 
upland valleys have the most severe frosts. Farther south, the former 
reputation of Buxton for severe frost owed much to the fact that the 
station was maintained in the bottom of a hollow surrounded by the 
bleak Derbyshire plateau, whereas the present position for the instru- 
ments is on top of a rise. 


Above the Tree Line: Sub-Arctic Britain 

Our knowledge of the high mountain climate of Britain owes most 
to the Ben Nevis Observatory built and maintained largely by the 
subscriptions of the Scottish Meteorological Society. Continuous 
observations were made from 1883 to 1904. Intermittent efforts have 
been made elsewhere; among the earliest were those of J. P. Miller of 
Whitehaven, previously mentioned as the founder of the Seathwaite 
rainfall record in 1845. He left minimum thermometers on Scafcll 
Pike and other summits; but they did not give trustworthy results. 
From 1937-1941 I kept a record on the summit of Great Dun Fell, two 
miles south-east of Crossiell in the northern Pennines. There is now 
(1961) Lowther Hill, 2,377 leet > ' n Lanarkshire. 

The chief characteristics of our higher mountain climate are 
increased wind force and increased duration of cloud cover; and 
all-pervading dampness is only too evident. On Dun Fell (2,780 ft.) 
the average wind force over the summit throughout the year 
appears to be about force 5 (20 m.p.h.). On Ben Nevis (4,406 ft.) 
the mean wind speed is upwards of force 6 (about 30 m.p.h.) 
and gales were extraordinarily frequent and often attained 
tremendous force. Above about 3,000 feet in Scotland snow is 
liable to fall in every month of the year, although even at 
4,000 feet it docs not often lie for more than a few hours between 
mid-June and late September. 

The higher and more exposed the summits are, die less is the 
average difference of temperature between day and night. On Dun 
Fell the mean daily range is in general less than half that of the low- 
lands; or Ben Nevis, less than a quarter. Excessively low winter 
minima do not occur, as on such occasions the lowest temperatures 
are found in the valleys; Ben Nevis never gave a minimum lower 
than i D . 

It will, however, be noted that the warmest month on Ben Nevis 
only averaged 41 °, and it is not surprising therefore to find only a little 
extremely hardy Arctic-alpine vegetation among the stones. Indeed 
with such excessive rain averaging 157 inches yearly, there is virtually 
no soil. On the Cairngorms at 4,000 feet (Plate 27, p. 190) the drier 
climate (less than half the rainfall of Ben Nevis) and greater proportion 
of sunshine allow more plant life. No one who wanders over our 


higher summits can fail to be impressed by the gallant struggle of the 
vegetation ; it should further be noted that shelter and aspect become 
very important at high levels. On favourable slopes facing the sun the 
shallow soil may warm up tremendously during the occasional anti- 
cyclonic spell, such as that which in June 1 939 gave several successive 
days with maxima over 70 on Dun Fell (2,735 ft-) where the normal 
daily maximum for June can be estimated as about 51 . 

Snow and snow-cover are considered in more detail in the next 
chapter. The small summit plateau of Ben Nevis is generally bare of 
snow, apart from remnants of drifts, from early June to the beginning 
of October; it can be regarded as "covered" on an annual average 
of about 215 days. The climatological snowline probably lies at about 
5,300 feet; that is, Ben Nevis would have to be about 5,500 feet high 
in order to harbour a small glacier to-day. Even in October and 
November snow-cover is rather intermittent. Warm air approaching 
from the south-west ahead of the normal Atlantic depressions is quite 
likely to give a temperature over 40 on the summit and if there is 
heavy rain the snow-cover is rapidly melted. After the mild and very 
rainy December of 1934 the summit carried very little snow even in 
early January. 

The average annual sunshine duration on the summit of Ben Nevis 
is one of the lowest yet observed anywhere; about 750 hours. Brief 
anticyclonic spells occur from time to time and a few hours of clear 
sky may follow each passing depression. Nevertheless the impression 
gained by anyone who has to sojourn for one purpose or another on 
our high summits is likely to be one of pitiless and nearly incessant 
raw cold wind and a great deal of low cloud, with abundant rime 
deposit throughout the winter. Even on a lower grass-grown summit 
such as Dun Fell the sheep do not wander up to graze before the end 
of May, and the run of the mean temperatures confirms that a growing 
season characterised by great day-to-day variability lasts only from 
late May to mid-September. 

The variability is the greater for plants at ground level, as they must 
accommodate themselves to cold wind, rain and lack of sunshine, 
with intervals of intense baking of the coarse, quick-draining shallow 
soils if sunny days occur; evaporation, too, in the exposed windy 
situation may be considerable for brief periods. And, while the winter 
temperatures are not excessively low, they are quite cold enough to 
inhibit any activity. Further, the snow-cover on the summits is often 



shallow owing to the severe drifting by wind, hence vegetation receives 
little protection. Elsewhere, the accumulation of drifts is such that in 
more sheltered locations the vegetation cannot quickly take advantage 
of any warmth, as the ground may be chilled by the melt-water for 
some lime even after the snow has disappeared. In the wetter parts 
of the Lake District the snowline to-day would probably be found at 
about 5,900 feet, and at 6,300 feet on Snowdon. Tables giving 
a summary of high-level data will be found in the Appendix. Note 
that the mean temperatures arc estimates based on the old 
observations, but brought for purposes of comparison to the period 
191 1-49. 

Altitude and Rainfall 

The increase of rainfall with altitude is a very well-known feature 
of our climate, arising as we all know from the additional uplift and 
consequent cooling of moist air-streams when they impinge against 
our high ground. But there are great local variations among our 
hills, some of which have already been mendoned. In the Lake 
District the convergence of the major valleys towards the central 
region, comprising the heads of Borrowdalc, Langdalc and the 
adjacent uplands, is a principal factor in the very high rainfall. For 
exactly similar reasons air-masses must often surmount the Crossfcll 
range and the Merrick hills in Galloway, of similar height; but in 
both these areas the fall on the uplands is about half as much as 
that which is measured in the central part of the Lake District. 

Farther norih, a small patch of south-west Inverness-shire at 
the head of Glengarry has probably over 200 inches a year. Yet 
hills near die sea of similar height, but slightly different trend, in 
Sutherlandshire appear to receive little more than half this amount, 
although gauges in these remote highlands are few. 

At altitudes up to 2,500 feet, wherever the slopes are less steep and 
the underlying rocks are impervious to water or nearly so, the result 

Plate 27 

Caihntoul and other summits looking S.W. from Cairngorm (4,084 feci), Aberdeen- 
shire, Banffshire and Invcrncss-shire: early June. Strato-cumulus and cumulus, 
fresh west wind. Snow beds in corrics, with remnants on plateau. Arctic vegetation. 
Excellent visibility characteristic of Highlands in this type of air. Temperature on 

summit 38 . 



B. A. <-roi«/l 

Cyril Xarhrrry 


of excessive cloud and rainfall in the pasi is seen in the thick mantle 
of peat. On the flat summits of the high Lancashire moorlands under- 
lain by resistant sandstones with nearly horizontal bedding, peat is 
found up to twelve feet or more in thickness in places where the 
present-day annual rainfall is of the order of eighty inches. This peat 
is the result of the decay of past vegetation under conditions of excessive 
rainfall and very poor drainage. Few plants can grow on such a cover, 
cotton grass and crowberry among them; heather prefers the patches 
of ground which are slightly better drained. 'Cotton-grass moor' as it 
is often called reminds us of the brief period in late June when even the 
most uncompromising Pennine moorland is dotted with the while 
blooms nodding in the wind. (PI. 18b, p. in). 

The varying aspect of our high moorland owes much to climatic 
factors. All the roads leading to Whitby over the lower and drier 
moorland of North-East Yorkshire, on which in general the rainfall 
is of the order of 38' are known to thousands for die magnificence 
of the display of heather at the end of summer. Elsewhere at inter- 
mediate levels with rather more rainfall there arc wide stretches of 
coarse upland pasture, difficult to reclaim and improve, on peaty 
soils often underlain by boulder-clay. Upper Wyresdale in Lancashire 
with a fifty-inch rainfall is typical; very wide stretches are found 
throughout the uplands of Central Wales, which is also a region where 
the prevailing annual rainfall is of this order of magnitude 

We have already shown that the upper limit of effective settlement 
is determined by the tree-line. The highest inhabited house in all 
Britain, and the highest reclaimed pasture, lie at 1990 feet flanking the 
Harwood valley which is tributary to Upper Teesdale. But we can 
also recognise that the reclamation and improvement of moorland by 
drainage is again limited, not by actual altitude so much as by annual 
rainfall. The limits of reclaimed pasture are found on Dartmoor at 
1,500-1,600 feet with about 65" of rainfall, except for the slopes in the 

Plate 28 

a. From Grasmoor: Biittermcrc and Scawfell Pike, Cumberland: fine afternoon, 
September. Characteristic flattened cumulus over mountains with westerly 
wind. Cirro-stratus spreading above. 

6. Raubol-rne, Derbyshire : unsetded wcadicr, July. Disturbed sky in the trough 
of a depression, showing cirrus, alto-cumulus and nimbc-siratus to right, with 
fragments of sirato-cumulus. 

C.B.S. ° 

I tj2 


immediate neighbourhood of Princetown where over 80' may be 
expected. Here additional labour has long been available. In the 
gritstone Pennincs, the limit appears to be set by about 6o", for example 
north of Bolton, in Bleasdale, on Pendle Hill, above Malham, in 
Mallerstang and in Weardale. On the Bcwcastle fells east of Carlisle, 
and in the Dumfriesshire valleys the profitable limit for reclamation 
is again of tliis order. Around Snowdon and within the Lake District 
it is broadly true that only those valleys which are exceptionally steep- 
sided and well drained have been setded beyond the limit of 60". 
Here and there, patches of limestone offer better drainage in parts of 
the Pcnnines with a high rainfall, for example above Dent. In the 
Highlands of Scodand the upland limit for reclamation appears to 
be set by a similar figure, except in a few mild steeply-sloping areas 
very close to the west coast. Throughout Britain occasional examples 
can of course be found of setdement in areas of excessive rainfall for 
different reasons. Blaenau Festiniog, for example has grown up on the 
slate-quarrying industry, and many hundreds of Welshmen accord- 
ingly endure a 97" rainfall. 

Reclaimed pastures mean habitations; and with regard to one of 
the most conspicuous features of British scenery — the wide stretches 
of completely uninhabited moorland — we may say that the determining 
factor is climate. Beyond a certain limit of rainfall the endless peat-hags 
on our moorlands are hopelessly unreclaimable; elsewhere, few plants 
will stand up to acid soils, cold and wind; and below the limit, economic 
factors have played, and continue to play, a large part. With a fifty- 
to sixty-inch rainfall most of Exmoor was profitably reclaimed a 
century ago, but it is a little uncertain whether similar efforts would be 
economically sound to-day. All those who walk in our upland districts 
will find abounding opportunity to correlate settlement and rainfall. 
Many who do not know upper Teesdale and Weardale will have had 
the opportunity of seeing another high-lying farm in Scodand. Four- 
teen hundred feet up in Perthshire lies Dalnaspidal; and as the 
Inverness express panted up the Drumochter on an October morning, 
many travellers during the war years have seen a pathetic field of oats, 
at best languidly ripening in a place with a July mean of 53 and fifty 
inches of rain. At a similar elevation in Teesdale we learn that two 
centuries ago a miserable bread was perforce often made from the 
oats cut before they were ripe. Farther north, after 'the black auchty- 
twa' — the fearfully late cold spring and wet summer of 1782 — it is 


recorded that the oats were carried in from the upland Banffshire 
fields in January. Such harsh seasons were long remembered. Urban- 
ised though we are, dependent on the bounty of other lands for much 
of our food, few intelligent Englishmen or Scotsmen can fail to note 
something of the annual struggle of the upland farmers against those 
chances of variations of temperature and rainfall whose effects we 
have seen. 

At this stage it may be asked, how far is the additional rainfall 
offset by better drainage, by aspect, and by the drying effects arising 
from increased exposure, to wind? The first question can only be 
answered empirically as yet. The inner Lakeland valleys owe their 
habitability to good drainage, as we have seen. Indeed as a result of 
the coarse soils and rapid run-off the effects of drought are quickly 
felt, and much more often than the southern visitor is prone to think. 
Similar effects can be found in die West Highlands. Elsewhere, the 
drift-covered slopes of many upland valleys are sufficiently free-draining 
to support quite a considerable farming population but no precise 
evaluation of the degree of slope or permeability of soil can as yet be 
given. It is significant however, that aspect plays relatively little part 
in our uplands, compared with the High Alps where the sites of farms 
and villages have been carefully chosen in the most sunny parts of 
favoured slopes; reference may be made to the exquisite diagrams in 
papers by Dr. Alice Garnett of Sheffield. In Britain the upper limit 
of reclaimed pasture is, in general, the same on either side of the 
valley; so that it would appear that in our cloudy and rainy climate 
free drainage is more important than direct exposure to the sun. 
Some studies have however been made which indicate that in a few 
localities such as Great Langdale the utilisation of the land by the 
valley farmers shows some recognition of the need to avoid as far as 
possible the shadows cast by neighbouring hills in winter. It has also 
been suggested that the successful arable farming on the Braes of 
Glcnlivct, at 1,200 feet north of the Cairngorms, owes much to 
the fact that this shallow upland basin is particularly well exposed 
to the light; moreover recent statistics suggest that here on the lee side 
of the mountains the average duration of bright sunshine is about 
10% greater than at Bracmar on Dccside. 1 Diffused radiation from 
the sky is also important for plant growth in northern latitudes. 

1 I am indebted for this information to Miss M. E. Charles and Mr. I. C. Rcid, 
during discussions in the Department of Geography at Cambridge. 



With regard to wind, a good drying wind in September is certainly 
greatly welcomed by the upland farmer anxious to harvest his oats. 
But exposure to drying wind on one day implies exposure to driving 
rain on another. On 20 September 1942, a field of oats, on a breezy 
and generally favoured south-west slope near Garrigill in Cumberland, 
was scarcely at any point other than green. At fourteen hundred feet 
above the sea, most of it was already beaten down by rain and wind, 
and on that day an unforgettable sense of hopelessness was reinforced 
by die weather; a gloomy sky with driving rain and a hard south wind. 
Too often in the past have such Septembers wrecked any prospect for 
the bold farmer who, encouraged by high prices and by no mean 
sense of patriotism however little expressed in words, has essayed the 
cultivation of even the most promising-looking breezy and sunny 
slopes at fourteen hundred feet — with a fifty-seven inch rainfall and 
scarcely twelve hundred sunny hours in a year, in the instance 
mentioned. On the Braes of Glenlivet the rainfall, after all. 
averages thirty-five inches and the limy drift soils add a material 

For upland position not only adds to the quantity of rainfall; 
the frequency and duration are also increased, though it must be 
emphasised that this increase is not in the same ratio as the quantity. 
Seathwaite receives five times as much rain as London, falling on 
about seven days for every five in the south. More precisely, measur- 
able rain falls on 236 days at the head of Borrowdale, 169 in London. 
Other representative totals include 207 at Falmouth, 196 at Cardiff, 
194 at Manchester, 217 at Greenock, 240 at Fort William and 263 at 
Stornoway in the outer Hebrides. Upland position means that there 
are a good many days in the year with one or two showers which the 
lowlands escape. Many good meteorological texts give diagrams to 
show that there will be occasions when the additional ascent of moist 
air over hills leads to instability and the growth of towering cumulus 
to a greater height than in the lowlands, hence a greater likelihood of 
showers. To such days must be added many others when the familiar 

Plate Xllla: Colliery smoke spreading and rising in turbulent daytime wind. 

b: November 1947: a rainy day in Manchester suburbs. The stand 
pipe had just been erected as a result of the prolonged laic summer drought, when 
domestic supplies throughout the city were about to be cut oil owing lo severe 

shortage of water. 

Plate XHIn. 


By permission of the Xational Smoke 
Abatement Society 

Tlie .Manchester Guardian 

P1.ATE XlVfl. 

C. H. Wood 

C. H. Wood 


persistent wetting drizzle associated with the low cloud and strong 
wind helps to swell the annual total. 

At a few stations automatic gauges record the number of hours 
during which rain of measurable intensity falls. Above Kinlochlcven, 
in the Argyllshire highlands near Fort William, the annual total of 
84 inches falls with measurable intensity on about 240 days and during 
1,550 hours, whereas in London the annual total of 24 inches falls on 
163 days during 437 hours. At an intermediate station (Eskdalcmuir, 
800 feet, in Dumfriesshire) the annual total of 56 inches falls during 
1,217 hours; at Southport 32 inches fall on 189 days during 685 hours, 
but at Armagh the same amount falls on 215 days during 848 hours. 
It will be seen that the total amount, the number of days, and the 
number of hours do not he in the same proportion. Most of the 
additional rainfall in our mountain districts results from more intense 
and prolonged rainfall on the wet days. 

With altitude, on the whole, low cloud, rainfall and wind force 
increase; range of temperature tends to decrease; average day-time 
maximum temperatures in particular decrease with height more 
rapidly than average minima. The amount of bright sunshine de- 
creases; and the length of the growing season, and rate of maturing of 
crops, decrease very noticeably. 

Some slight advantage is gained, it would appear, with regard to 
thunderstorms. Over most of our highland districts they appear to be 
rather less frequent than in the lowlands, although at intervals they 
may produce considerable devastation due to the rapidity of the 


Here and there our northern hillsides show the scars of long past 
torrents due to sudden violent downpours. One of the most appalling 
visitations of this type befell on Stainmore in 1930, and is described in 
British Rainfall for that year. Brast Clough, on the western flank of 
Pendle Hill in Lancashire, is said to have been largely eroded by the 
deluge in a violent thunderstorm in the sixteenth century. 

Snow, again, is an element whose frequency increases very rapidly 
with altitude. So large a part in our impression of the scenery of 

Plate XI Va: A characteristic cloud-sheet over Lancashire in a humid air stream 
from S.W. round the margin of an anticyclone. Smoky patches are visible in the 
cloud. Warm subsiding air above, hence little or no upward growth of cumulus. 
b: Liverpool and its smoke seen dirough a gap in die cloud-sheet. 


Britain is played by snowfall in some parts, and by lack of it in others, 
that the subject deserves a chapter to itself. 


Chapter ro 

Buchan, A. and others (1905-10). The Ben Nevis Observations. Trans. 

Roy. Soc. Ediiib. Vols. 42-44 (also many articles in J. Scot. Mel. Soc.) 
Garnett, Alice (1935). Insolation, Topography and Settlement in the 

Alps. Amer. Geogr. Rev. 25: 601-17. 

(1939). Diffused light and sunlight in rcladon to relief and settle- 
ment in high latitudes. Scot. Geogr. Mag. 55: 271-84. 
Manley, G. (1942). Meteorological Obscrvarions on Dun Fell. Q.. J. Roy. 

Met. S. 68: 151-65. 

(1943). Further Climaiological Averages for the Northern Pennines. 

Q,. J. Roy. Met. S. 69: 257-60. 

(1945). The Effective Rate of Aldtudinal Change in Temperate 

Adantic Climates. Geogr. Ren. 35: 408-17. 
Pearsaix, W. H. (1950). Mountains and Moorlands. London, Collins' 

New Naturalist. 
Salisbury, E. J. (1939). Ecological Aspects of Meteorology. Q.. J. Roy. 

Met. S. 65: 337-57. 
Tansley, A. G. (1939). The British Islands and their Vegetation. Cambridge. 

University Press; indispensable; gives abundant references for further 



Since 1950, in addition to the station at Moor House (1,840 ft.) now 
under the Nature Conservancy, upland observations have become available 
at Malham Tarn Field Centre (1,300 ft.), at Onecotc (1,350 ft.) in Stafford- 
shire, at Bwlchgwyn (1,297 ft.), Alwen (1,100 ft.) and Llyn Stwlan (1,656 
ft.) in North Wales. Observations near the summit of Ben Machui were 
also kept for a few mondis by Dr. P. Baird, then at the University of 



The belching winter wind, the missile rain 
the rare and welcome silence of the snows 

R. L. Stevenson: To My Old Familiars 

. . . l/te uniformly scandalous 
Condition of the snow! 

Climber's Ditty 

Over a large part of the British Isles the more impressive extremes 
of our winter weather occur with somewhat dangerous rarity. 
Further, their effects on our consciousness tend to be mitigated in 
towns and cities. To this we must add the fact that a rapid improve- 
ment in the amenities and case of transport and in the efficient heating 
of public buildings has taken place during four decades, 1898- 1939, 
characterised by a predominance of mild winters. 

Eighty per cent, of our people live in towns; hence many of our 
people are inclined to forget that snow is still a very important factor 
in our northern uplands, perhaps more so on account of the very wide 
variations in its occurrence, whether in amount, persistence, or season 
of year. Snow came back noticeably into the Londoner's consciousness, 
for the first time for several years, in December 1938. Since then the 
severe winters of 1940-41-42, the cold January of 1945, the very severe 
February of 1 947 and the cold December of 1950 have affected all Britain. 
February 1954 and 1955 were snowy; February 195G severely cold. 

It remains to be seen whether, in 1980, we shall continue to de- 
scribe winter extremes as dangerously rare. Such a phrase may yet be 
advisable when we consider how little provision is made. The stupen- 
dous absurdity of our plumbing arrangements, dating from a time when 
the revenue of private water companies was considered to be more 
important than the comfort of householders, could only be tolerated 


in a climate which for several years allowed the drivers of the early 
locomotives to stand, like coachmen, inconsiderately exposed to the 
weather. The awful January of 1838 called attention to the need for 
improvement. Our tendency to hope for the best with regard to frost 
and snow may owe a good deal to that element in us which for cen- 
turies endured windowless huts and a smoky fire — such as we still 
see in the west of Ireland. The Iron Age camp on the summit of 
Inglcborough testifies to a curiously able people (compare the Des- 
borough mirror) with an astonishing endurance of cold and damp, 
who carried their scorn of Continental heating arrangements righi 
down to the eighteenth century. Some of those who had gone to 
America were then gradually seduced from their old allegiance; and 
to meet the harder winters of a new continent Benjamin Franklin 
devised his stove. At last the New Englander was able to resist the 
charms of bundling. 

But their cousins retain much of their ancestral endurance, having 
less need for technical improvements unless die British climate takes a 
markedly colder turn. With regard to frost, snowfall and transport 
problems February 1947 probably did good, in reminding us that 
tolerance of out-moded institutions may be carried too far. 

Average temperatures being what they are (January, from 45 to 
39 at sea level) every part of the British Isles must expect an occasional 
winter day with sleet or snow rather than rain. In an earlier chapter 
we saw that in the cooler months air-masses cither of polar maritime or 
polar continental origin may reach us with a surface temperature even 
on the coasts below 40° ; but as they travel over an open sea it is rarely 
that the wind from seaward gives a surface temperature below freezing- 
point at coastal resorts. Hence unless a snow-cover is very deep and 
1 he wind persistently off the land the ground adjacent to the coast will 
rarely remain covered for more than a few hours. Even the Scilly Isles 
however are occasionally snow-covered, when an exceptionally severe 
outbreak of cold air from the Continent reaches the islands as a north- 
easterly gale from the direction of the nearest land tliirty miles away. 
Perhaps the most unpleasant experience those normally mild islands 
have had for fifty years past was the gale of 29 January 1947, when the 
maximum temperature at Plymouth was 23 and even at the Scilly 
Isles only 28°, in spite of their being surrounded by sea-water with a 
surface temperature of nearly fifty degrees. February 1955 again gave 
a heavy drifting snowfall in West Cornwall. 



For snowflakes to be observed it appears that the freezing level 
should not be more than about 2200 feet above the observer. If 
the British Isles were entirely flat we might expect the frequency 
of days with snow or sleet falling to increase steadily northward and 
eastward, as a result of the greater proximity and frequency of arrival 
of air-masses sufficienUy cold at the surface to allow snow to reach the 
ground. More often than not it can be observed that on a day with 
continuous precipitation, i.e. saturated air, from low cloud in winter, 
rain only will fall if the air temperature is 39 or above. At 38 an 
alert motorist may see an occasional blob of sleet among the raindrops 
on his windscreen as he drives; at 36 a few flakes falling in the chilly 
rain will be seen by walkers; and at 34 largish half-melted snowflakes 
will predominate. Note that if the surface air is not saturated, snow- 
flakes in a passing shower might thus reach the ground with an air 
temperature as high as 43 , for example on cold, windy but sunny days 
in late April; this comforms with our experience. 

But if the temperature falls to the freezing-point, fine dry snow- 
flakes will drift before the wind, and the lower the temperature, the 
finer the snow as a general rule. 

Eastbound drivers leaving Exeter on a sleety winter morning soon 
appreciate these facts as they approach the Somerset uplands. Even 
in outer London, the climb from Croydon to Woldingham provides 
abundant experience. 

The rapid diminution with temperature in the characteristic size 
of snowflakes leads to an increasing tendency to drift, and to penetrate 
into houses. Hence we find that drifting snowfalls occur quite rarely on 
low ground near the sea, but that they are a regular feature of our 
higher mountains. Further, from what has just been said with regard 
to rain, sleet and wet snow at temperatures above freezing-point it will 
at once be evident diat on many days when nothing more than chilly 
rain is observed in our lowland cities, heavy snow will be falling at 
the 2,000 feet level. 

Altitude then combines with latitude and proximity to colder air 
supplies to increase the frequency with which snow falls. Moreover, 
just as in the last chapter we saw that the presence of hills and moun- 
tains tends to increase the frequency of rain as well as the amount, we 
can now see that precisely similar results will apply with regard to 
snow. Snow showers are very commonly a consequence of instability 
in Arctic or Continental air reaching our shores. Cold dry Continental 



air leaves Denmark with a temperature of perhaps 20 ; after crossing 
the open North Sea its surface temperatures at Tyncmouth or Spurn 
Head may be 34 . But the result of the warming, in combination with 
the increased humidity, is not only an extraordinarily unpleasant 

rawness; the surface air is 
thoroughly unstable, and on 
meeting any obstacle giving a 
slight upward push it will rise, 
forming cloud and vigorous 
showers. At the temperatures 
named these are sometimes 
merely sleet and hail on the 
coast, but a few miles inland 
on high ground they are of 

Nothing is more character- 
istic of our north-cast facing 
coasts than the driving snow- 
showers accompanying the 
onset of Arctic air throughout 
the winter months. These are 
particularly noticeable where 
the coast rises steeply, as 
in Cleveland, or where high 
hills rise but a few miles inland, for example, on the Lammcrmuirs. 
Dwellers in Norfolk will recall how often they break on the morainic 
hills behind Cromer and Sheringham, and farther south, the rela- 
tively high frequency of snow on the Downs in East Kent, by 
comparison with the hills nearer London, derives largely from the 
same cause. Instability snow showers due to the arrival of cold air 
over a relatively warm sea also occur from time to time elsewhere, 
in Eastern Ireland and, with a north-wester, in such places as the 
North Wales coast. 

We can see the combined result of these factors in the map 
(Fig. 58), which shows for stations at low levels the average frequency of 
days per annum on which snow or sleet has been observed to fall, 
taken over the period 1912-38. In the Scillies and in the far south-west 
of Ireland the average is three or four days only. In the Outer Hebrides, 
although the average temperatures are but little lower, proximity to 


F10. 57 
Average frequency of snow cover in 
relation to the mean temperature o( 
winter months: based on nine upland 




SO 100 ••* ■ 


4- 0* 

Fio. 58 

Average annual number of days with snow or sleet observed to fall (after the author's 

paper in Meteorological Magazine, 1940, by permission of Her Majesty's Stationery 

Office). The data refer to ground below 200 feet. For higher levels add to die figure 

shown by the map one day for each 50 feet above 200 



colder air supplies results in an annual frequency of 25 days at Storno- 
way. Towards the east of England there is a gradual rise to between 
15 and 18 days in Norfolk and over 20 in Yorkshire. Additional 
instability due to the presence of hills gives rise to greater frequency 
in Cleveland, the Lammermuirs and Aberdeenshire; at Aberdeen an 
average of 34 days yearly with snow or sleet is observed. 

It must be stressed that the map below is representative of places 
less than 200 feet above the sea. To the figures shown by the map one 
day should be added for every fifty feet of elevation above 200 feet. 
If this is done the frequency at the highest villages in the N. Pennines, 
about 1,500 feet, is seen to be about 50 days yearly. Similar figures 
apply to the higher farms in South Scotland and to places in the High- 
lands such as Braemar, Tomintoul and Dalwhinnic. For higher 
levels we can note that on the summit of Ben Nevis the average annual 
number of days with snow or sleet observed to fall was 169 during the 
twenty years that the observatory was open. 

Snow and snow-cover are elements of climate whose frequency is not 
so easily observed as some might imagine. The Meteorological Office 
has gradually standardised the manner of recording ; a 'day with snow' 
is recorded on any day when snow or sleet is observed to fall, however 
small in amount. From what has already been said it is evident that 
not all observers will be equally alert with regard to the occasional 
'blobs' of sleet falling in cold rain; and at stations such as lighthouses 
and airfields where a watch is kept throughout the night, the observed 
frequencies tend to be slightly higher than at 'climatological' stations 
where the instruments arc read once daily and the observer is not 
always present. Allowing a little for diis, since 19 12 the observations 
over the country as a whole have been sufiiciendy consistent for maps 
to be drawn, such as that which follows (Fig. 59, p. 204). 

A day with 'snow-lying' is recorded when at the 9 a.m. observation 
the visible country surrounding the station, at the same level and 
characteristic of the station itself, is more than half covered by snow. 
It is evident that in the milder locations, and early and late in die 

Plate XVa: Seathwaitc, the Stye and Great End; fine April morning (1910), a 

few winicr snowdrifts above 2000 ft. The wettest place in England; annual rainfall 

182 inches by the farm, 185 inches locally on the fell beyond. 

b: Ski-ing on Ben Lawers. Typical fine windy winter day; snow slightly 
drifted, grasses showing. 

Plate XVa. 

G. P. Abraham, Ltd. 

Tilt Timet 


1 ' 



season, such a definition does not imply a 'day's duration of snow- 

Broadly however, it can be said that the number of 'days' duration' 
is about the same as the 'frequency of snow-lying at 9 a.m.' whenever 
the mean temperature for a month is less than 37 . In what follows it 
will be convenient to refer to the annual frequency of days with snow- 
cover; but it should be remembered that as observations are at 9 a.m. 
an annual average of 1-3 at Torquay, or 2*1 at Eastbourne does not 
necessarily mean that snow lies throughout one or two whole days. 
But in the Scottish Highlands at Braemar where the annual average 
(1913-1949) at die 9 a.m. observation is 66-4, the number of days 
during which to a casual observer most of the ground by the village 
is covered throughout is probably very close to this figure. 

The number of days of snow-cover depends largely on the mean 
temperature, which is chiefly a question of altitude; and partly on the 
frequency with which snow falls, partly on the quantity. We have 
already seen that frequency is greater, on account of relief, in certain 
areas towards the north-east coasts. Correspondingly, frequency of snow- 
fall is less on the lee-side of those hill ranges over which northerly and 
easterly winds must descend. One of the most snow-free areas of the Scot- 
tish mainland lies at Prcstwick; in England, the head of Morecambe Bay 
is similarly favoured. Further, as most of our heavy snowfalls arrive on 
winds from an easterly point, when depressions are centred to the 
southward, we find that there is a very marked orographic influence. 
Just as the south-western flanks of our mountain districts lead to rapid 
increase in rainfall, the easterly and north-easterly slopes are almost 
always associated with rapidly increasing depth of snow whenever there 
is a major fall. Exceptions occur now and then, for example when the 
dominant wind direction is southerly and a stationary or slow-moving 
front lies north to south in the Irish Sea. Such a situation in 
January 1940, January 1947 and again in March 1947 gave a 
particularly heavy snowfall in the Lake District. Occasionally too 
the rather rare development of what is called a polar-air depression 
gives a heavy snowfall attaining its greatest depth in somewhat 
unexpected places, for example north-west Ireland in April 191 7 

Plate XVIa: Snow on the beach at Perranporlh in Cornwall, December 1938. 

b: Floods in Wharfedale. Winter floods following heavy rain are often 
reported from die middle courses of many of the Pennine rivers. 




100 50 30 20 10 5 ° 

Days of srvow ly»r\q 



Fro. 59 

Number of mornings with snow lying, annual average, 1912-38 (after the 

author's paper in Meteorological Magazine. 1947- by permission of Her 

Majesty's Stationery Office) 


and the Lancashire Pcnnincs in May 1935. (Sec also Figj 36, 
p. 113, 10 May 1943). 

In general however the regions of heaviest snowfall arc well 
marked. Throughout the eastern highlands of ScoUand snowfalls tend 
to be heavier than farther west; Dccsidc for example is more snowy 

than Spcyside. In England 
snowfalls are most common 
in the eastern Pcnnincs, 
and at high levels they 
can be very persistent. 
Sheffield's high-lying sub- 
urbs are well-known in this 
respect; also the efficiency 
of its transport system. 
H uddersficld is more snowy 
than Oldham; Consett in 
Durham and Quccnsbury 
in the Wcst-RidingofYork- 
iest urban communities in 
Britain. Nearer the coast, 
die wide windswept York- 
shire Wolds repeatedly 
figure in our newspapers; 
again, the amount falling 
is generally considerably 
greater than in the plain 
of York to the leeward, or 
in low-lying Hull. The 
Lincolnshire Wolds are prominent for the same reason. The snowy 
sections of the Great North Road are well-known. The Stamford- 
Grantham section is a mere four hundred feet above the sea, but, 
rising like the Wolds, from the plain to eastward, it has a heavier 
snowfall. In 1947 the hold-up of road transport here was remarkable; 
similarly, in 1708 Ralph Thorcsby was forced to spend five days at 
Stamford on account of deep snow. Other uplands characterised by- 
heavy snowfall against which easterly winds are forced to rise include 
the hills of north-cast Wales, East Kent, and Dartmoor. Dartmoor 
snowfalls arc occasionally extremely heavy and are the more noticeable 

Fio. 60 
0700 hrs.. 25 February 1933. Very heavy 
orographic snowfall in N.E. and N. England 
and in N.E. Ireland associated with a low 
which moved westward ; snow 6 inches deep at 
Durham but 30 inches deep in Tccsdale at 
800 ft. (For notation, sec p. 5) 



in the relatively mild south-west; again we can see how on stceply- 
nsing moorlands facing the damp unstable east wind as it sweeps 
down-channel from the cold Continent not only the quantity of snow, 

but the associated drifting 
at higher levels can be very 
great. With regard to trans- 
port the position is not 
helped by the ease with 
which the deeply entrenched 
Devonshire lanes flanking 
the moor can be blocked. 

Convergence of the sur- 
face air-streams also plays 
its part with regard to 
regions of excessive rainfall, 
as we saw in the Lake 
District. Similar factors 
operate with regard to snow- 
fall; in a south-easterly 
snowstorm for example the 
upper glens of Angus into 
which the air-streams 
converge, receive very heavy 
falls. Convergent air-streams 
also affect Dartmoor snow- 
falls; with an east wind, very 
exceptional depths were 
reported near Holne in 1929. 
Upland snowfalls may occur 
during many months of the 
year. During the past half-century the earliest noteworthy fall was 
that which covered the Pennines above about 800 feet on 20 Sept- 
ember 1919; while an exceptionally late fall, nearly a foot in deplh 
in some upland Lancashire towns, occurred on 16 May 1935. Some 

F10. 61 

1R00 lira., 20 February 1941. Towards the 

end of the great N.E. coastal snow storm, 

depth 30 inches in Newcastle and 42 inches 

at Durham, 18-20 February (see p. 5) 

Plate 29 

Hartjnoton, Derbyshire: brilliant noon in the Peak District, March. Snow slightly 

glazing on surface, air temperature 28 . Cold anticyclonic day with gentle north 

wind, slight growth of cumulus over the distant plain. 




representative figures with regard to frequency of snow and snow-cover 
are given below, together with maps; sec also Table VIII in the 

Estimates for higher levels of the average monthly and annual number of 
days with snow-cover prevalent during die last three decades. 


F M A M June to Sept. 

O N D 


N. England 
(i, 840ft.) 


17 17 6 1 o-i 

2 6 12 



Ben Nevis 
(4,406 ft.) 


28 31 30 20 g 

14 22 30 


Taking all the available evidence it appears that everywhere in 
Britain at higher levels the overall increase with altitude in the average 
annual frequency of snow-cover is nearly linear. In ScoUand it can 
be put at about I day for 22-25 feet. Further south data are more 
scanty but the increase is probably not so rapid; about 1 day for 
30 feet is a fair estimate for South Wales. It appears that at levels up 
to 800 feet the range of variation between very mild and very severe 
winters lies from about one-quarter of the average figure (zero where 
the average is less than 10) up to about fifty days above the average. 
Even from a small scale map it will be seen that in spite of the appre- 
ciable chance of snow in September and May virtually no part of the 
British Isles below 2,000 feet has on the average as many as 100 days 
with snow-cover. Throughout the winter warm rain, such as accom- 
panies the onset of maritime tropical air (or even, at times, returning 
mP), is capable of removing practically all traces of snow. 

From this it is evident that at any inhabited level snow-cover is 
neither regular nor continuous in its occurrence. We can judge llus 
very well from the figures for Bracmar (1,120 ft.) where snow can be 
expected to lie for about half January and February. During the past 

Plate 30 

The Cairngorms and the Lairig Ghru from Rothicmurchus, Invcrnrss-sliire: early 
June. Anticyclonic weather with clear skies above; strong southerly wind giving 
cloud on the mountains and clearer skies on the descent (partial "fohn"). The 
lowest visible snow patches are at about 2,700 feet. 

CB.S. p 



37 years with an average of 66, the range of variation in the numher 
of days with snow-cover has lain from 34 (in 1938) to over 122 (in 1919). 
Nearby at Balmoral (930 ft.) the figures are 61, 27 and 116. At 
Braemar there were twelve winter months during which snow has lain 
throughout the month in the village; these include 1 January, 8 Feb- 
ruaries, and 3 Marches. By the time we reach levels at which an 
average of 100 days with snow-cover can be expected, we can be pretty 
certain that in an average year ground will be found covered from mid- 
December to some time in March, except for brief periods. 

Hence it might be argued that ski-ing accommodation might 
profitably be established at this level (2,000 to 2,200 ft.) where indeed 
the chances of snow-cover in Scotland are about equal to those at the 
lower Alpine resorts. Unfortunately, the conditions for ski-ing arc 
rendered the more difficult, not only by the weather; terrain plays its 
part. Damp wind, mist and succeeding frost create appalling surfaces ; 
as a result of fusion they arc commonly made up of large and thor- 
oughly uncomfortable crystals, and may be so hardened on the surface 
that control is difficult. Drifting moreover occurs to such an extent 
at higher levels that the depth of snow is extremely variable. (PI. XVb, 
p. 202). Icy patches interspersed with projecting boulders, followed by 
irregular drifts hardened by the incessant wind often provide challeng- 
ing difficulties even on a sunny day, which are hardly conducive to the 
relaxed enjoyment offered by a high-altitude resort in the central Alps 
within the European winter axis of high pressure. And unless one lives 
in Edinburgh or Glasgow the cost of the journey is still considerable. 
Nevertheless many of those who live near our higher hills know how 
to take advantage during winter week-ends of the fine interval fol- 
lowing a fresh powdery snowfall. Nothing is more welcome than such 
a day of opportunity seized and held within the Scottish winter, when 
under the low January sun in a wedge of high pressure, the opal- 
escence of the polar air lights up the slopes of the Central Highlands 
above Dalwhinnic or Loch Tay with astonishingly beautiful pearly 
colouring. Opalescence is probably a by-product of salt particles 
suspended in the lower atmosphere after stormy weather; it appears 
to belong only to high latitudes, for example Iceland. Further south, 
there are the rare occasions when the snow-covered Kinderscout 
peat-hags dazzle the Manchester man's unaccustomed eye, and only 
the dreadful brown smoke-haze penned within the surface inversion- 
layer over the plains to westward remains as a reminder of the murky 


gloom in the city. Such a day was 5 January 1941; in Westmorland 
minima in the Eden valley fell to zero, and in the clear air above the 
valley inversion the rosy flush of dawn on snowy Crossfell was unforget- 
table. On the same morning the minimum at Barton, outside Man- 
chester was also o°, but at Oakcs on the ridge above Huddersfield 
(760 feet) 1 9 ; compare Ushaw and Houghall in County Durham, 
page 167. 

Indeed the significance of snowfall in Britain lies in the further 
variety of scene it provides, often at most unexpected times and places. 
Occasionally among our hills a fiercely brilliant May sun from a 
cloudless sky blazes in Alpine fashion on a deep snow-cover, such as 
covered the Isle of Skye almost to sea-level on 2 May 1923. (Cf. PI. 29, 
p. 206). At other times, large corniced drifts have been seen flanking 
the Great North Road within a few miles of London. Cuttings hall 
a mile long in the snow at the head of Weardale were eight to twelve 
feet deep when, after a seven weeks' blockage, the first car passed over 
Killhope summit on 2 April 1937. In December 1938, the Times 
published a photograph of the snow-covered sands of Cornwall ter- 
minated only by the Atlantic (PI. XVIa, p. 203). 

Snow is capable of giving a remarkable amount of trouble in this 
normally mild country. Even the innocent monotony of the outskirts 
of Cambridge has provided its story. In February 1799 Elizabeth 
Woodcock floundered into a drift on her way home from Cambridge 
market to Impington three miles away, and was buried for eight days 
before she was rescued — still alive — after hearing the Histon church 
bells twice. Farther north each snowy winter is generally marked by 
tragedy when keen but unprepared walkers lose themselves in snow- 
storms on the moors. When an exceptionally snowy season befalls, 
such as February 1947, catastrophic losses of sheep must often be faced. 
Taken over the country as a whole, February 1947 was probably 
unmatched by any other month since January 18 14 for persistence of 
cold, dullness and repeated heavy snowfalls. In the London area, 
statistics of occurrence of snow can be assembled since 1668. 

Permanent Snow 

We have already seen that, on high ground above 1,000 feet or so, 

drifting is characteristic of the majority of heavy snowfalls. Roads, 

especially where they run between walls, arc only too easily blocked, 

and if the resultant depth of snow is greater than about eighteen inches 



they must be dug out. Sometimes weeks of labour arc needed, as in 
Tecsdalc in March 1937. Drifts in the lee of walls and other sheltered 
places may persist long after the country as a whole is free from snow. 

As a result it is common to see occasional drifts here and there 
among our mountains, especially in gullies facing north and east and 
thus sheltered from sun (and driving rain; continuous warm rain is 
perhaps the most effective agency for removing a snowfall). In a 
normal year the last drifts at the 2,000 feet level in the N. Pennincs 
linger till about the middle of May. In the northern gullies of Ilcl- 
vellyn (2,800-3,000 feet), towards Red Tarn and Brown Cove, some 
small patches of snow may be found till early June; and the same is 
true of the Great End Gullies above Borrowdale, and the northerly 
crags of Carncdd Llewelyn in North Wales. 

Rarely there have been reports of strangely persistent drifts in 
exceptional places at much lower levels. In 1947 in a shady location, 
remnants of a drift in the Northern Cotswolds were visible until mid- 
July, and Bonacina has reported that in the same district in 1634 a 
snowdrift is said to have survived until August. Nearby, the village 
of Snowshill, eight hundred feet up with a northerly aspect, is likely 
to owe its name to past experience, reminiscent of Cold Ashby in 

Persistent snowbeds occasionally linger in the full sunshine; follow- 
ing a north-easterly blizzard in March 1937, snow accumulated 
just under die summit plateau of Crossfell on the south-west side. In 
spite of a sunny May, the drift remained visible until 10 June. After 
the terribly severe March of 1785 followed by a warm May, one of 
our first Lake District tourists commented on the large snowdrifts on 
the Crossfell range still visible from Penrith on 18 June. 

There is no doubt that the lingering snowdrifts of winter add their 
quota to the charm of the mountains in spring. In April in die Lake 
District the fresh green of the larches around Grasmerc (however 
much they were detested by Wordsworth) is the more noticed when 
die northern slopes of Helvellyn or Bowfcll still carry much snow, 
liven more is this true of the higher Scottish mountains. Above 3,000 
feet, on the Cairngorms and Ben Nevis, on Mam Soul, Ben Wyvis 
and other high summits large snowdrifts linger until July, and in some 
recesses through August. In two exceptional locations they normally 
jusi last through the year, the critical period being the last half of 



These semi-permanent snowbeds both lie at about the same level, 
3,750-3,800 feet, on the north-cast flank of Ben Nevis and on the same 
side of Braeriach, in the sheltered and deeply shaded gullies descending 
steeply below the summit ridge (PI. XVIIIb, p. 227). Forty years ago 
these beds were judged to be permanent; and it is on record that some 
considered that the degree of consolidation of the snowbed on Ben 
Nevis almost justified the view that it was an incipient glacier. But 
alas for the enthusiasts; the amelioration of the North-west European 
climate in the last four decades has been reflected, in Scotland, in the 
total disappearance for a brief period in certain years (1933, 1935, 
1938, l 945, 1953 and 1958) of these drifts, towards the end of 
September. It is interesting to note that in each year in which the 
beds failed to survive, a warm and rainy autumn and a winter of 
little accumulation was followed by a warmish summer, rainy in the 
latter part (i.e. August). 

Eighteenth-century travellers in the Highlands espied the snow- 
drifts on the Cairngorms (August 1727) and on Ben Nevis (September 
1787), while Pennant in 1771 believed those of Ben Wyvis to be per- 
manent. For Helvellyn in the Lake District there is some interesting 
evidence that in the Napoleonic era early last century snowdrifts were 
more persistent than we should nowadays expect. John Dalton was a 
man of very regular habits, and year by year made the ascent about the 
end of the first week in July, at the beginning of his annual holiday 
from Manchester. He was fond of determining the humidity of the air, 
and to do this he cooled a small vessel if possible to the dew-point. 
For this purpose melting snow was convenient. In several years 
between 1805 and 1823 he records that he found snow remaining "in 
the usual place, about a quarter of a mile north of the summit," that 
is overlooking Brown Cove. To find snow remaining there nowadays 
so late in the year would be most unusual; but in 1812 a large drift 
was still to be found on 18 July (1951, 10 July). Late in June 1817 
Dorothy Wordsworth noted that snowdrifts remained above Red Tarn; 
no doubt her brother's line about the recess "that keeps till June 
October's snow" derives from this excursion. Protracted cold springs 
were rather frequent during the years in question. In mid-July 1843 
Miller noted a remnant near the summit of Scawfell Pike; in August 
1 818 Keats crossed snowbeds as he walked up the west side of Ben Nevis. 
Studies of the extent to which our average temperatures of summer 
have varied in the past 200 years suggest that after exceptionally cold 



springs and summers it is just possible that a rare snowdrift might 
survive through the year in the N. Pennincs and the Lake District, 
and perhaps even in North Wales. Odd stories to the effect that 'some 
say the snow never melts up there' have been heard in the north of 
England, and were probably handed down from the occasional cold 
year in the past such as 1695. That such stories were current long ago 
is shown by an account of an ascent of Crossfell in 1747 when the 
writer emphasised that "on August 13 in spite of careful inspection of 
all the likely places, no trace of snow could be found." 

Such stories indeed remind us that there is always a tendency to 
romanticise with regard to such extremes as snow. Nevertheless, it 
does not do to neglect the possibilities of snow, whether as an element 
in scenery, an impediment to comfort, or a dangerous menace to 
survival. The experience of 1 947 is too fresh in memory. In the last decade 
the winter of 1950-51 was marked by exceptional persistence of moun- 
tain snow-cover above 2,000 feet; and for the first time for many years 
snowdrifts in the Lake District and North Wales lasted far into July. 
February 1954 and 1955 also gave heavy snowfalls. 

Calculations based on die average temperature of the summer 
months lead to the conclusion that in the regions of heavy precipitation 
the present day snowline would be found at about 5,300 feet in die 
Ben Nevis region, 5,900 feet in the Lake District and 6,300 feet in 
the Snowdon District. That is, if the mountain summits exceeded 
this height small glaciers would probably develop under present climatic 


Chapter 11 

Ashmore, S. E. (1952). Records of snowfall in Britain. Q_. J. Roy. 
Met. S. 78: 629-32. 
Bonactna, L. C. W. (1927). Snowfall in the British Isles, 1876-1924. 
British Rainfall, 67: 260. 

(1936). Snowfall in the British Isles, 1925-1936. British Rainfall, 
76: 272. 
Manley, G. (1939). On the Occurrence of Snow-cover in Great Britain. 
Q_.J. Roy. Met. S. 65: 2-26. 

(1940). Snowfall in Britain. Met. Mag. 75: 41-48. 
('947)- Snow-cover in the British Isles. Met. Mag. 76: 1-8. 
(i949)- The snowline in Britain. Geogr. Annaler, 31: 179-93. 



The cold earth slept below 
Above the cold sky shorn 
And all round with a chilling sound 
From caves of ice and fields of snow 
The breath of night like death did flow 

Many readers of this book will not lack acquaintance with the story 
of the rocks and the climatic conditions under which they were 
laid down. For an introduction to this subject others may well be 
commended to the companion work in this series on Britain's Structure 
and Scenery by Professor Dudley Stamp. Therein they will be reminded 
of the evidence that two hundred million years ago, and to a less extent 
thirty million years ago, the climate of this country over endless ages 
by our present time-scale was warm and humid to a degree which we 
should nowadays associate with lower latitudes. Such climatic varia- 
tions have indeed affected the character of the rocks we see; but are 
more properly considered in a work on geology. 

We should rather concern ourselves with the more recent climatic 
variations that have affected the surface aspect of the country, the 
vegetation and through them the life of animals and men. The period 
of man's existence and development covers the past million years. 
During that span of time those parts of the earth in which man has 
most notably evolved have been subject to the exceptional climatic 
vicissitudes culminating in the several phases of the last Ice Age. In 
the British Isles the effects of the glaciation on our landscape are so 
widespread and conspicuous that it would be convenient to begin 
any discussion of secular variations with an account of the factors 


governing the greatest climatic variation oi all, which took place but 
yesterday by comparison with the geologist's time. Indeed for the past 
ten thousand years the country as we now sec it has been slowly re- 
covering from the effects of the last phase of the Ice Age. Even in 
comparison with the million years of man's evolution this period is 
very short, but in it all the developments of the civilisations we know 
have been comprised. 

We must first try to envisage the state of affairs before this series of 
wide climatic oscillations began. In the earlier Tertiary period it 
appears that the world was nearly, if not entirely, free from ice. 
Geological evidence from deposits of that period in Spitsbergen and 
the Antarctic indicates that the seas were open while on land, even 
in high polar latitudes, considerable vegetation was then to be found 
which we should now regard as typical of more temperate countries. 
Weather was stormy and disturbed; and the vigorous air movement 
now characteristic of the Icelandic seas prevailed in winter throughout 
the Polar Basin. 

Later in the Tertiary very extensive mountain-building movements 
took place, culminating in Europe in the uplift of the greater part of 
the Alps and associated southern European ranges. Towards the Poles 
extensive uplift of existing land masses such as Greenland and Scandi- 
navia was associated with a correspondingly widespread degree ol 
foundering in the North Adantic, and with the great volcanic out- 
bursts of which Iceland is the principal remnant. 

It has been shown that even with an ice-free Polar Sea the winter 
temperatures, though very similar at die North Pole to those we 
experience in Britain now, were such that snow would intermittently 
fall on the neighbouring land. The uplift of high mountain ranges 
adjacent to the Polar Seas, therefore, would not only lead to an in- 
crease in precipitation; at higher levels much would fall as snow, and 
gradually accumulate. Evidence is still rather controversial with 
regard to the exact period at which sufficient accumulation of snow on 
a high land-mass such as Greenland existed to initiate an ice cap. 
Undoubtedly, however, there should be a gradual spread of 
melt- water, and floating ice, from Greenland's marginal glaciers; this, 
being less dense would form a sea-surface layer above the denser salt 
water — as it still tends to do at the present day. Further, in 
winter such a surface layer would freeze again much more easily 
dian if it were salt 


Gradually then we observe the chain of events; high land in high 
latitudes; ice caps within the marginal mountain ranges from which 
glaciers descend towards the sea; melt-water spreading over the sea, 
favouring as a whole die slow increase of sea-ice winter by winter. 
Once die sea-surface is extensively frozen, it assumes to a large extent 
the properties of a snow-covered land surface. Radiation on clear 
winter nights leads to cooling of the adjacent surface air, and the slow 
building up of the resistant "cushions" of cold air that we recognise as 
a characteristic of the present-day Siberian winter. In an earlier 
chapter we named them cold anticyclones, and we noted their pro- 
pensity to fend off die surface incursions of warmer air. 

In his standard work Climate through the Ages, Dr. C. E. P. Brooks 
has shown by ingenious argument that whereas over long periods of 
the earth's history the polar seas were just warm enough to be open, 
if such an open polar sea began to freeze, it is probable diat over a 
period of years the ice would gradually increase in area until its edge 
lay about io° from the Pole. Consequent upon the existence of such an 
extensive icy plain winter temperatures in the surface air would fall 
to far lower values than formerly prevailed. Gold winds spreading 
outward from this cap of Polar air would be likely to engender much 
more vigorous depressions along the Polar Front between them and 
the warmer maritime air to southward. In turn, however, the resulting 
stronger winds near the margin of the Arctic would tend to break up 
the floating ice, so that the sea-ice would not continue to spread 
indefinitely towards the equator. 

Meanwhile we must recall that in the Pliocene or late Tertiary 
Scandinavia also stood higher than at present; with die wider spread 
of cold air and the increased storminess the accumulation of snow on 
the Norwegian and Swedish highlands began. Ice-covered highlands, 
like ice-covered sea, create dieir own cold. The reflexion of solar 
radiation from a snow surface is so great, amounting to about 80% 
of that which arrives, that it is not difficult to envisage the steady 
development of an ice cap from a stage where not merely scattered 
snowdrifts, but many square miles of ice, resisted the sun, warm wind, 
and rain of summer. 

All these processes owe something to the changes in the elevation 
of land in high latitudes relative to the sea. It is probable that the 
continents were also more extensive. It can be shown that in higher 
latitudes an increase in the degree of continentality, that is, of the area 



of land surrounding a given station, will lead to a greater fall in the 
winter temperatures than can be compensated by the rise in summer. 
If for example, the present-day Baltic became dry land the net result 
would be a fall in the overall average temperature of Sweden. Without 
doubt we can build up a tolerably satisfactory explanation of the 
establishment of vast ice-caps in Greenland, over Scandinavia and 
over eastern Canada based on the greater extent and elevation of land 
in high latitudes towards the end of the Tertiary. Fringing the great 
Scandinavian ice sheet the mountains and highlands of Britain were 
also heavily glaciated and from them as most of us know, the ice 
flowed southward towards and across the English lowlands, almost as 
far south as the Thames valley. 

For the detailed results with regard to scenery reference is better 
made to Professor Stamp's companion volume. It will serve here to 
remind the reader of the widespread mantle of glacial drift; the 
diversion and subsequent modification of many of the rivers, here 
rejuvenated, elsewhere incised in attractive gorges such as that at 
Durham, at Ironbridge, or above Llangollen. Our mountains arc of 
relatively low altitude but they are nevertheless true mountains ; in this 
respect they surprise many visitors from abroad. Peaks and ridges, 
screes and crags, corrics and their tarns all testify to the effects of 
extensive and severe glaciation. So much so, that it is at times difficult 
for the climber to believe that once above the rocky lip of the corrics 
in Skye there will not after all be a small remnant of the ice, however 
convinced he is from his reading that such remnants were last visible 
about 8300 B.C. 

But the Ice Age, represented for us by a vast development of ice 
over Scandinavia and by extensive glaciation of our own islands, was 
not a single episode. Within the last million years or so it is probable 
that the great ice-caps of Scandinavia and Eastern Canada waxed and 
waned in at least four major phases, accompanied to a varying degree 
by corresponding advances and retreats of the mountain glaciers of the 
Alps and other ranges, and by minor fluctuations of their peripheral 
extent. Between these major phases of the Ice Age, intcrglacial periods 
of considerable length occurred and in some districts there is evidence 
that the climate became milder than at the present day. Deposits of 
intcrglacial age have been identified in the Alps in which the plant 
remains indicate a more southern type of vegetation than is permitted 
by the present temperatures. 


Tliis problem of multiple glaciation is one of the world's most 
fascinating puzzles. Those who would be prepared to accept an 
explanation based entirely on the results of the elevation of land in 
high latitudes find it very difficult to conceive that Scandinavia could 
bound up and down, so to speak, several times in such a short period 
having regard to the earth's history as a whole. Other suggestions have 
been made which postulate considerable variations in the intensity of 
solar radiation. It has been demonstrated by Sir George Simpson 
that a small increase in the power of the sun would ultimately give rise 
to increased cloud and precipitation in highland regions towards the 
poles; assuming that the land was already sufficiently elevated, the 
resultant increased cloudiness and snowfall would gradually give rise 
to an ice cap. He points out the importance of the fact that a wide- 
spread cloud sheet, once formed, reflects a great deal of the radiation 
falling upon it. The elegance with which his theory can be extended 
to explain the occurrence of cooler and warmer interglacials is attrac- 
tive; it was published in the Quarterly Journal of the Royal Meteorological 
Society for 1934, with some revision in 1957. But unfortunately, 
sufficient geological evidence is not forthcoming with regard to the 
relative coolness or warmness of the several interglacials which Simp- 
son's theory would require; interglacial deposits are rare, as they are 
generally removed by the succeeding glaciation. For this reason the 
elucidation of the full story of the British glaciations is tardy. Moreover 
full agreement has not been reached with regard to the number and 
extent of the several glaciations in other parts of the world. 

Other theories with regard to the cause of widespread refrigeration, 
as some have called it, rest on small variations in the intensity of solar 
radiation arising from slight changes in the tilt of the earth's axis and 
the eccentricity of the orbit. The effects arising from such causes arc of 
disputable magnitude. Then, apart from variations in the behaviour 
of the sun itself, the amount of solar radiation transmitted through the 
atmosphere to the earth's surface may have varied on account of 
changes in the gaseous constitution of the atmosphere. Such changes 
at low levels appear rather unlikely; we are as yet uncertain of the 
possibilities higher up. Another suggestion arises from our knowledge 
that terrestrial volcanic eruptions result from time to time in such an 
explosive outburst of fine dust, that the climate of the whole world is 
affected due to the slight curtailment of solar radiation reaching the 
earth's surface. But although certain historic cold years are known to 



have been so caused (1784, 1816, 1912 for example) the finest volcanic 
ash does after all sink back to earth within two years or so of the erup- 
tion. To explain an ice age lasting a hundred thousand years would 
demand eruptions on a scale which again appears incredible for other 
reasons. It cannot be denied however, that vast eruptive disturbances in 
combination with several other factors might, so to speak, have pulled 
the trigger, had they occurred at the right moment. But until we have 
satisfied ourselves that the glaciations in various parts of the world were 
contemporaneous we cannot answer this and many other problems. 

Puzzling features of the glaciation problems keep appearing. For 
many years it was thought that we were definitely living in the waning 
of the last Ice Age; barring some slight evidence of a slight recession 
towards greater cold about 500 B.C. Since about 1920 it has been 
recognised that the climate of much of the north temperate zone 
took a turn towards greater cold within the last seven hundred years. 
As part of this comes within the period of organised science and in- 
strumental recording, we will discuss the details later in this chapter. 
We must leave the questions of why such a significant climatic fluctu- 
ation has occurred as one part of the greater question — that of multiple 
glaciation, for whose solution the combined efforts of workers in a 
variety of sciences will for long be required. 

Let us rather endeavour to reconstruct the climate of the Ice Age 
in Britain at its maximum. Evidence for such a reconstruction is 
forthcoming from a variety of sources. First, we know that there must 
have been floating ice adjacent to our S.W. coasts, and that many of 
the icebergs melted but a short distance from land. This is shown by 
the many erratic boulders dredged from the sea bed. Banks of shelly 
drift near the East Anglian coast testify to an earlier stage in which, 
while the North Sea was still unfrozen, very strong and persistent 
easterly winds prevailed. Thirdly, floating ice debouched to the north- 
west of Scotland, and we must therefore allow for a broad ocean to the 
west of Britain. For very long periods however most of the North Sea 
was a vast plain covered by slowly moving ice from the Scandinavian 
Highlands, indicated by the many erratics of Norwegian origin found 
on and near our cast coasts. Similarly, the shallow Irish Sea, the North 
Channel and the channels among the Hebrides were completely 
covered by the great tongues of ice, or piedmont glaciers, spreading 
from the adjacent highlands. For convenience the map from Britain's 
Structure and Scenery is reproduced on the next page. 


We have to devise 
a scheme which al- 
lowed the survival of 
forest trees in North- 
ern France, although 
Southern England 
was a bare windswept 
tundra snow-covered 
for a great part of 
the year. The Eng- 
lish Channel was 
merely a wide gulf 
extending eastward 
from the Atlantic. 
We have to devise a 
pressure distribution 
to allow for certain 
features elsewhere; 
in Bermuda the sand 
deposits indicate an 
excess of strong south- 
westerly winds in gla- 
cial times. Further, 
if there were so great 
an amount of float- 
ing ice in latitudes 
south of Iceland and 
west of Britain, we 
may assume that 
the surface waters of 
the North Atlantic 
owed much to the 
melting of the ice 
and hence that they 
were much colder 
and presumably less 
saline than now. 

Polar air from W. to N.W. reaching Britain would do so over a 
very much colder sea than at present. In such air, through the greater 

Fio. 6a 
The geography of the period of the maximum glnria- 
tion in the British Isles. This map shows the approxi- 
mate position of ice sheets at the second glaciation. 
The home ice-caps — centres of ice accumulation and 
dispersal — arc numbered; lines and arrows show 
direction of ice flow; where lines are broken the 
earlier directions were superseded by those shown by 
solid lines (by courtesy of L. D. Stamp) 



part of die year snow showers would prevail even at sea level. We can 
see our analogue in those present-day South-Adandc islands which arc 
annually beset for a time by pack-ice, such as South Georgia or even 
the South Orkneys. South Georgia in 54°S. has a mean temperature 
for the warmest month of 43 °, and for the coldest 29 . 

Eastward and north-eastward the great ice-cap prevailed over 
which the surface air could at best be warmed little above the freezing- 
point. But not far from our shores the salt Adantic water remained 

unfrozen and indeed it 
seems likely that off Portu- 
gal its temperature was 
not much below that of 
the present day. 

Hence we can arrive 
at a scheme of winter and 
summer weather. Milder 
Adandc air would meet 
the polar air very com- 
monly just to the south- 
west and south of Ireland. 
Deep depressions following 
this polar front would give 
very heavy precipitation 
in the south-west, but 
during much of the year 
rain would often fall 
rather than snow. Even 
in winter, with a warm 
open sea but a few miles to the southward, rain would occur now and 
tiicn sufficient to wash away much of the snow on low ground; the 
like occurs nowadays in extreme S.E. Greenland. We might thus 
hazard a climadc explanation for die remarkable survival of certain 
relatively delicate plants in S.W. Ireland, while not far away behind 
the Killarney mountains to the north-eastward much of the country 
lay under ice. For the mountainous south-western promontories 
would be almost constandy under heavy cloud ; and in clearer weather, 
almost incessant stormy and damp winds would hinder the ponding 
of the cold air close to the coast. At the present day, by the gloomy 
fjords of western Tierra del Fuego, stunted evergreens survived in an 

0600 hrs., 6 March 1947 


appallingly cloudy, wet and chilly region in which it is nevertheless 
rare to record more than ten degrees of frost. A hundred miles to the 
eastward, in the lee of the mountains, the climate is far drier and 
much colder in winter; persistent lowland snow-cover and tempera- 
tures around zero are quite common. 

Similar results would be found in Cornwall and Devon. The onset of 
milder air from the not-distant Bay of Biscay was evidendy frequent 
enough to prevent the winter accumulation of snow on Dartmoor from 
persisting sufliciendy to establish glaciers except perhaps on a very 
small scale. Yet, a hundred miles further north, the glaciers from the 
Brecon Beacons were sufficiendy powerful to reach the Bristol Channel. 
The mild damp air from the seaward, flowing over the fringe of icy 
waters near our coasts and then across the barren cold land would give 
heavy low cloud and drizzle for much of the summer in Cornwall, 
Devon and along the south coast. In winter, the very sharp contrast 
with the cold air inland would be marked by vigorous depressions in 
which heavy cold rain, sleet or snow would fall depending on the 
distance from the warm sea water. An almost exact working model of 
the conditions of the Ice Age can be seen in the events of February 
1947, in south and south-west England. (Cf. Fig. 63, p. 220). In the 
events which culminated in the great Midland snowstorms of 5 March, 
in place of the expected thaw whose northward travel did not extend 
beyond the Thames, we can see why the limit of the ice sheet in one 
glaciation almost reached the Thames, while in later advance it 
scarcely spread south of Yorkshire. For thousands of years the battle 
between tropical and polar air was repeatedly fought out over Southern 

Farther north almost endless snowstorms would be expected in 
winter along the western fringe of Britain associated with depressions 
making their way up the west coast along the chilly half-open seas. 
So marked, however, were the maritime effects that even in the Outer 
Hebrides the slopes of the higher hills appear to have been free from 
ice; and to have allowed the survival of a number of hardy plants. 
Farther to the eastward these coastal snowstorms would be less in- 
tense, and with distance from the open sea the weather throughout 
the year would become appreciably drier and more sctded, with 
some sunshine, especially in spring and early summer. As a result 
the quantity of snow farther to the eastward was not sufficient to cover 
so much of the high ground throughout the year. In summer, parts of 



the Peak and North-East Yorkshire stood above the ice as broad, 
barren, frost-shattered uplands with occasional exceptionally hardy 
Arctic plants. Farther soutii, although the Downs were free from 
glaciation there is still some reason to suppose that in the shady 
hollows facing north-east persistent snowdrifts remained throughout 
most years. Some consider that such snowdrifts played a part in 
sculpturing the hollows by 'nivation', not only there but along the 
Marlborough Downs and perhaps elsewhere. Such coombs arc clearly 
to be seen along the scarp of the South Downs near Eastbourne. Not 
all authorities, however, are agreed on this point. 

Other evidence agrees with the meteorological reconstruction, 
according to which few depressions would penetrate to north-central 
Europe with sufficient energy to give heavy rain. Throughout Central 
Germany and even in North-East France the widespread 'loess' soils 
derive from lite immense clouds of fine, dusty soil raised in a dry 
climate by the cold north-easter flowing outward from the ice-cap, 
especially in spring. Just as in the Fenland to this day the chilly 
north-east wind of spring, flowing over the warm black soil on a sunny 
morning, becomes turbulent as a result of surface heating and raises 
clouds of the fine soil, so in the past; for most of the phenomena of the 
Ice Age we have to-day admirable working models. One difference 
between conditions then, and the initial stages reproduced in Feb- 
ruary 1947, lies in the much greater snowfall of the eastern flanks of 
our uplands in the north compared with the west. This characteristic 
feature of modern snowy winters arises from the presence of the broad 
North Sea; in glacial times such a sea was not in existence. The 
greatest snowfalls probably occurred tliroughout our western highlands 
in a manner reproduced for us by the events of 29 January 1940 and 
12-13 March 1947. On the former date the snowfall along the western 
flanks of the Pennincs and the Lake District was so heavy that the 
southbound L.M.S. express was held up for 36 hours a few miles south 
of Preston. In mid-March, 1947 a rather similar "south-wind snow- 
storm" affected much of north-west Britain, especially on the 
uplands. It may be added diat a minor event of the same type befell 

Plate 31 

Loch Achtriociitan, Glcncoc, Argyll: August evening. Intermittent sunshine 

with patches of characteristic rattier disturbed strato-cumulus and cumulus at 

3,000-4,000 Tcct among mountains. 


I- LATE 31 



on 6-7 January 1947; with a nearly stationary front running N-S, 
through the Irish Sea, the south wind on its eastern side gave about 
nine inches of snow in Borrowdale, but very litde east of Penrith, or 
on the Pennines beyond. 

Estimates have been made of the extent to which temperature in 
England at the height of the Ice Age was lowered compared with the 
present day. In the extreme south the July average was probably in 
the middle forties. Farther north the presence of such a large ice sheet 
would imply that at sea level in Eastern Scotland the summer averages 
were a few degrees above freezing-point, while in winter they may have 
lain between io° and I5°F. At the latter season severely cold clear 
weather and stormy cloudy weather with a much higher temperature 
alternated; the increased amount of cloud and wind towards the 
west of Scotland would there lead to milder winters, just as at present. 
It seems likely that the difference in average winter temperatures 
between, say, Edinburgh and the Hebrides was of die order of three 
times that which prevails to-day. Yet even with regard to Eastern 
Scodand it must not be forgotten that a largely open and very stormy 
sea lay at no greater distance and separated by a smaller mountain 
barrier than that lying between Ostcrsund in central Sweden and 
the Norwegian coast, across which the present-day January tem- 
peratures differ by 12 or thereabouts. 

From the meteorological standpoint diere would also be a con- 
siderable improvement in the weather, away from the Adantic. Deep 
depressions would be found travelling eastward towards the Bay of 
Biscay, giving in Northern France heavy precipitation, often as rain. 
The existence of a massive surface layer of cold air over most of the 
British Isles would ensure that many warm fronts in winter would 
remain nearly stationary near die south-west coasts. The mild 
Adantic air ascending over the cold surface air would give snowfall 
much more frequently than rain in districts such as the English 

But at intervals other depressions would pass northward along the 
partly open sea west of Scotland. The air in their warm sector would 

Plate 32 

Westmorland: June afternoon. Normal fair-weather strato-cumulus and cumulus 
of turbulent humid westerly current in early summer, ascending over hills. Kent 

estuary in foreground. 
C.B.S. 0_ 



however be considerably chilled by its northward journey over the 
cold icy waters, or up the ice-covered plain where we now find the 
Irish Sea. Probably many of these depressions would rapidly decline 
in vigour, although die frequent sweep of the cold moist south wind 
along the western isles of Scotland and beyond to the coasts of Norway 
would still, as we have seen, give frequent and heavy snowstorms and 
very extensive low cloud. Following the northward passage of such 
depressions cold west to north-west winds would prevail with frequent 
snow showers on our mountainous coasts; a situation resembling that 
at the present day, save that the air was probably ten to twenty 
degrees colder. We can imagine ourselves sitting on the Donegal 
mountains in June. For miles around we should see the gently- 
sweeping contours of the ice, covered by melting snow at the higher 
levels, grey and rather dirty-looking below. Overhead, a broken sky 
of ragged snow-laden cumulus with hard blue gaps; here and there in 
the distance the curtains of blinding snow showers passing over the 
higher hills, but falling in wet and heavy flakes on the greyish bare ice 
around Londonderry. Towards the Atlantic the dark water sky from 
which the raw west wind blew would be seen; eastward, the while 
glare on the horizon would remind us of the clearer skies and sunshine 
extending over hundreds of miles of lifeless waste, only broken by the 
dark streaks of rock on the sunny slopes of the Isle of Man, a few 
struggling lichens perhaps on Crossfell, and beyond the bare stony 
warmth of the hills of North-East Yorkshire from which even the 
gleaming Cheviots might often be seen refracted upward on days of 
calm clear air. The photograph (PI. XVIIIa, p. 227) of a retreating 
glacier on a cloudy day in Central Norway has been chosen as we may 
expect that something very similar would be found occupying the upper 
Pennine valleys during the waning of the last phase of the Ice Age, 
about 15,000 years ago, and rather later in the valleys of the Scottish 
Highlands. There would also be many days of widespread low cloud, 
such as we often now find covering the gloomy basaltic fjords of North- 
west Iceland. 

Nothing indeed fires the imagination more than the evidence of the 
immense range of variation in the effective climate of these islands over 
a relatively short space of time. In countries such as Norway small 
glaciers still exist and the ice age does not appear so remote. In the 
lowlands of France there was no glaciation at all, and although pine 
and birch replaced the deciduous forest of the present day, there was 


still abundant life. But in Britain the change to a temperate climate 
to which an astonishing range of plants can be adapted from one in 
which almost the entire realm was lifeless, is the more startling. Yet it 
agrees well with the evidence of our eyes; that under British climatic 
conditions a very short journey is needed to produce a remarkably 
effective change in the environment. No definition can be given of 
such a climate; no precise classification can apply. Suffice it to remind 
ourselves that even to-day, although many exotics flourish in mild 
coastal locations, our tree-line lies at a mere two thousand feet. To find 
a similar contrast it is necessary to climb six thousand feet or more in 
the Alps or the Rockies. 

Post-Glacial Chances 

Having reviewed the circumstances of the greatest of the climatic 
vicissitudes of which we have evidence we might now consider the 
minor, but none the less significant variations under which the evolution 
of the present inhabitants of the British Isles has proceeded. Primitive 
varieties of the present human stock certainly existed in England in 
the intcrglacial period between the last major advance of the ice and 
the last but one; they may have been here in the previous interglacial 
period, retreating to warmer lands during the tens of thousands of 
years through which each ice-cap persisted. But from many points of 
view wc can consider that the slate was virtually wiped clean when, 
about 20,000 years ago, the final retreat of the last great phase began. 
Southern England and the Midlands were then and for long remained 
a tundra; ice had not spread quite so far as in previous glaciations, but 
except perhaps in shelter in the extreme south the climate was cold 
enough to prevent the growth of trees other than the dwarfed Arctic 
birch and willow creeping here and there along the ground. Hence 
practically all the vegetation and the animals we now know (including 
the more evolved varieties of man) have made their way into Britain 
within the last 15,000 years. 

Within the great barren tundra-heath lying to the south of the 
retreating ice-sheet drainage was often poor and immense hollows 
existed in which the slow growth and decay of boggy vegetation led to 
the formation of peat. As the climate gradually became warmer trees 
spread. First there came the hardy tree-birch followed by the re- 
sistant pine, both of which we still find advancing up the valleys 



towards the highest levels at which trees will grow in Norway, and 
northward towards the shores of Lapland by the Arctic sea. The 
pollen from the trees growing in the neighbourhood of bogs and of 
small lakes was carried over diem by the wind, and some deposited. 
Now, by microscopic analysis, the number of grains of pollen repre- 
sentative of several different kinds of tree at any given depth in the 
bog, or lake-bottom deposits, can be counted, hence the proportion 
of different trees in the surrounding woodlands at various times during 
the past ten thousand years or so can be estimated. This ingenious 
technique, largely developed since 1921, has given us for most of 
Northern Europe with die British Isles the character of die plant 
succession and hence the general nature of the climate. For a pre- 
dominance of birch and pine indicates diat while the winters were still 
cold, summers were relatively warm and dry, such as we now find ai 
2,000 feet in eastern Norway. At slightly higher levels towards the 
surface of the bogs the proportion of pine increases further, then it 
decreases and the proportion of alder pollen becomes large, together 
with hard woods such as oak and elm. Without doubt this indicates 
a change to a considerably wetter climate with milder winters. 

In some parts of Denmark (Allerod) and locally in South Britain 
at least as far north as Berwickshire the late-glacial and post-glacial 
deposits reveal a narrow layer containing a little tree pollen between 
wide layers which were laid down where the landscape was devoid 
of trees. This warmer phase is often known as the "Allcrod oscillation" 
and appears to have lasted for upwards of a thousand years. From 
this and odier evidence we are led to assume that the amelioration of 
climate following Glacial times was not at all regular. We can imagine 
die tide of vegetation advancing northward as the summers grew 
warmer, but then being thrust back for some hundreds of years before 
the advance was resumed. Recent work by Winifred Pennington on 
the sediments of L. Windermere has led to an approximate dating 
in agreement with the Swedish geochronology; roughly 10,000 to 
8,850 B.C. for the warmer phase and 8,850-8,200 B.C. or thereabout 
for the reversion to a colder climate. It appears probable that this 
was the period when the last small glaciers descended into the heads 
of the Lake District valleys, leaving their moraines for example near 
Stocklcy Bridge on the path from Seathwaite to the Stye. 

To a certain extent those climatic fluctuations can be correlated 
with Other events. The increased dampness accompanied by much 



i i 

3 S 

" S b-5 - — 

3 Q. 



2 O 

s « 


.3 5 


S I 

o u o 
3 o 

w CO 

§ 2 

.Q ^ 

O ~ 
o « 

" ^ c c 

£ 5 
o a 
Z 5, 

2 2- 

a u 3 



■P H. 

1 1 

Z c 

c c 

^ M 

O « - 

be rt 

C r- 

l § 


a = 


3 >• u ,-• 

2 S 

« .o 



. -* » — 'i 






a ^ = -~,-c 


- s 
£ Si 


milder winters, shown by the rapid and widespread increase of alder, 
oak and elm, accompanied the changes in the extent of land and sea. 
A general rise of sea level was associated with the opening of the Baltic 
to Atlantic water and a considerable enlargement not only of the 
Baltic but also of the early North Sea. These events are placed by 
recent workers about 6,000-6,200 B.C. and readers of Professor 
Stamp's work in this series will recall his suggestion that the rapid 
amelioration of climate owed much to the fact that the circulation of 
salt ocean waters round Britain became freer. For at some time during 
diis period the evidence indicates that the Straits of Dover were also 
finally opened. But as the general rise of sea-level was largely the 
result of the world-wide melting of the great ice-caps it is not easy to 
disentangle the primary cause of such extensive and rapid melting, 
remembering how long the ice had been in existence and the inter- 
mittent character of the subsequent climatic recovery to which the 
earlier Altered oscillation testifies. 

A rough time scale is given in the table. 

summary table 
Variations of Climate in Britain During the Past i 5,000 Years 

Before 14,000 B.C. 

Be/ore 1 1,000 B.C. 

After 10,000 B.C. 

About 8,850 B.C. 

Ice Sheet slowly retreating over Scandinavia and 
N. Britain with halts and brief readvances at times. 
Tundra and heath throughout S. Britain, birches 
slowly advancing in S.E. Climate cloudy, windy 
and raw, with minor oscillations. 

Spring and early summer relatively dry. (July 
mean temperature about 45°, Midlands). Winder- 
mere becomes a lake after retreat of ice. 

'Allered phase'. Climate appreciably warmer. 
Tree birches reach Berwickshire, Lake District and 
Central Ireland. (July in the Midlands about 54'). 
Occasional permanent snowdrifts among moun- 
tains, but probably no glaciers S. of Scotland. 

Deterioration. Birches retreat S.E., tundra in 
N. England (July— N. Midlands— about 48°-50°). 
Re-establishment of small glaciers in Lake District, 
Wicklow and N. Wales; extension in ScoUand (the 
'Highland Readvance') lasting about five centuries. 



Following 8,200 B.C. 

Improvement to 'Boreal'; birches spread N. followed 
by pine; summers steadily warmer; sea becoming 
warmer off British coasts. N. Sea still largely dry 
land, winter severe at lirst, milder later, but still 
drier and colder than today. Weather probably 
more anticyclonic than now. Elm and oak spread 
later in period: summers quite warm at end. Final 
disappearance of remaining mountain glaciers. 

About 6,000 B.C. and 
'iibsequenl milknia 

'Atlantic' phase; sea level rises considerably, North 
Sea broadens; possibly final opening of Straits of 
Dover. Summers remain warm but ground 
generally more moist, winters milder than to-day. 
Rapid and extensive spread of oak, alder and also 
lime replacing pine and birch. Climate gradually 
attains a degree of breezy warmth comparable with 
warm, moist and cloudy years (1834, 1852); July 
perhaps 65 in N. Midlands and January in lower 
forties. Mountains decidedly wet arid windy; trees in 
Highlands not found quite so high as in subsequent 
drier phase. 

About 3,000-2,500 B.C. 
and onward 

'Sub-Boreal': in the north, drier and less windy. 
Winters drier with some frost, summers rather 
warmer than at present, equally rainy in S. Birch 
and pine prominent again on drier areas, also up 
mountains to about 3,000 feet in Central Scotland, 
but not so high nearer coasts. Upland settlement 
widespread, glaciers in Central Norway probably 
disappear. Decided fluctuations at intervals but not 
apparently of great length until :— 

About 500 B.C. 

Climate again much damper with considerably 
cooler and more cloudy summers, less evaporation, 
more wind and rainfall. Rapid growth of peat over 
previously forested uplands especially where less 
well drained. Tree-line lowered by perhaps 1,000 
feet. Birch increases in lowlands and in damp 
sites oak, alder and willow especially prominent. 
Summers perhaps 4 cooler than previous phase, 
winters still rather mild due to much wind and 
cloud ('early Iron Age'). 




Possibly minor amelioration and recession in 
Roman times; improvement about 7th and nth 
century, wetter around 1100, again more disturbed 
after 1300. Minor fluctuations with tendency for 
colder winters after 1550; tendencies probably 
more or less similar to those shown by Fig. 64. 
Minor drier and wetter groups of years in S.E., 
but uncertainty how far these are applicable in 
N. and W. Prevalence of colder winters in late: 
17th century, and recurrence 1740 onward; groups 
of generally warm summers, e.g. 1772-83; and cool, 
1692-1700, 1 809- 1 8. Tendency in direction of 
milder winters since 1850 or earlier but not unin- 
terrupted. Appreciable increase of average tem- 
perature in spring, summer and autumn since 1930. 
Despite 1959, the peak may have been passed. 

Post-glacial forest history in Britain pursues a broadly parallel, 
but by no means identical course in different areas. The general 
picture however, is clear; that of the march to and fro of different 
types of forest as a direct response to changes of climate. 

A climatic optimum was reached upwards of 2,500 years B.C. since 
when in spite of partial recoveries there have been several small but 
significant steps towards greater coolness in N.W. Europe. Perhaps the 
most notable of these steps is represented by the deterioration which 
set in about 500 B.C. The intervening slight ameliorations of climate 
in the direction of greater warmth and dryness have been interrupted 
by further slight recessions and if we accept the Scandinavian dating 
as being broadly applicable to much of Britain we may put the reces- 
sions at about 400 and 1300 A.D. It is however probable that the 
effects on temperature and rainfall in these last fluctuations were not 
greater than those we now experience between groups of warmer and 
drier years, and later groups tending to more unsetded and windy 
summers with more cloud and less evaporation. Minor fluctuations 
of this type lasting for a decade or so can be recognised even in the 
past two centuries of instrumental recording. 

These very significant variations within the period since Neolithic 
times arc important. On what do they depend? That we cannot say 
as yet; perhaps a clue to die problem will eventually be forthcoming 
when the smaller fluctuations within the 'instrumental' period have 



been fully analysed. For although the variations have not been large, 
even within the last 250 years short spells have occurred which, if 
they persisted, would be likely to affect the natural vegetation and it 
is a matter of deciding how such spells might become more lasting. 

We can summarise the effect of these climatic fluctuations on the 
habitability of the British Isles. At a time when the earliest civilisations 
were well developed in lower Mesopotamia and Egypt, between 4,000 
and 3,000 B.C., we can imagine a decidedly damp and mild Britain 
heavily forested except in limited hilly areas where exposure to wind. 
better drainage and the character of the soil encouraged a more open 
vegetation. Mesolithic cultures may still have been represented by 
scattered groups living here and diere towards the sea-coast, for 
example, on the limestone uplands of East Durham. Neolithic peoples 
were about to arrive, possibly across a narrow strait of Dover, or making 
their way along the Nordi Sea and Adantic coasts. Among other things 
they brought with them was the art of pottery; and the distribution ol 
the fragments they left gives us a good indication of the routes they 
followed towards the more favourable areas for setdement. 

Slowly the climate became drier, probably not without interruption ; 
and the dense tangled woodlands throughout the clayey valleys foi 
long remained an obstacle. Fifteen hundred years later wc find the 
evolved men who built Stonehcngc, who quarried for flint so success- 
fully in sandy, dry West Norfolk, who for some time had cultivated 
grain and were beginning to know of the use of metal, still living on the 
uplands and on the naturally-drained areas where trees were more 
scattered. In Northern Britain the evidence indicates that about 
2,000-1,500 B.C. the climate was drier than at present. Examination 
of the peats indicates that again the pine and hazel were spreading; 
and other evidence which it is tempting to relate to climatic factors 
can be found. In Dorsetshire the recent excavations at the great 
ramparts and earthworks of Maiden Castie show that for several cen- 
turies it was occupied as a pre-historic fortress-site by a relatively 
accomplished people until a period indicated by the finds of bronze 
weapons as somewhere about 1,500 B.C., perhaps two centuries or so 
after the erection of Stonchenge. Then it was suddenly abandoned, 
and left unused and untenanted for twelve hundred years, until about 
300 B.C. the obvious advantages of such a defensible site again appealed 
to later Celtic-speaking tribesmen of the Iron Age. Now it is tempting 
to think on the dry chalk downs die abandonment of the site was 


enforced by the lack of sufficient water; as the winter climate became 
drier, the water-table in the underlying chalk would fall below that 
to which wells could be dug. That the summer temperature and rain- 
fall was not seriously changed in these extreme southern parts of England 
is however suggested by other evidence. During the Maiden Castle 
excavations under Dr. Mortimer Wheeler large numbers of charcoal 
specimens were examined dating from Neolithic times, the Iron Age 
and Roman times. The width of the annual growth rings docs not 
differ significantly from that of present-day representatives of the same 

Yet, at about the same period, the presence of Irish gold ornaments 
in Scandinavia is attested, and points to the development of a good deal 
of traffic across the North Sea. It is argued that such navigation in 
the very primitive vessels of the time would only be practicable if die 
weather was quieter than we should now expect. The meteorology 
of the period might plausibly be explained by analogy with the situa- 
tion during occasional dry years such as 1887, 191 1, 1921, or better 
still, August 1947. The Azores anticyclone, well to the north of its 
normal position and extending far to the north-east, would give dry 
quiet weather with fight variable winds throughout Northern Britain 
and across the North Sea to Norway. But in Central Europe and also 
in Southern Britain thundery summer rains would still contribute, as 
they do to-day, an appreciable proportion of die annual rainfall. 

As the vegetation of Central Europe depends largely on the summer 
rainfall we should not therefore expect such marked evidence of 
drought, even in S. England, as might for example appear in north- 
west Scodand. 

If during the winter months the Azores High still tended to occupy 
a much more northerly position than at present, Southern England 
would again be decidedly drier at that season than now. Wc may seek 
an analogy in the phenomenally dry, mild February of 1932. For the 
replenishment of the chalk in Britain it is the winter rainfall that is 
necessary; and the phenomena of the Bronze Age in Britain and Scandi- 
navia might reasonably be explained on the assumption that the Azores 
High spread farther north both in winter and summer than at present 
and was more persistent. Icelandic depressions probably were less 
vigorous, and also moved on tracks well to the northward. Such a 
situation might well arise if dierc had been a considerable decrease in 
the amount of ice in the Arctic Seas. This again would favour Western 



Scandinavia in that winters would be less stormy; and the greater 
dryness would undoubtedly be of benefit to Ireland. Further the 
dryness of Ireland at this period is supported by the evidence; in a 
Donegal peat bog twenty-six feet beneath the surface, relics of a well- 
built log hut dating from Late Neolithic times, were discovered 
some years ago. Elsewhere, the growth of good timber in areas now 
completely covered by peat supports the view that the Bronze Age 
climate was not only drier, but less windy dian now. Within the peat 
flanking the summit of Crossiell, stumps are found indicating that at 
some time trees grew up to 2,500 feet above the sea; and in Central 
Scotland trees arc known to have reached 3,000 feet. Yet from the 
nature of the vegetation growing in Bronze Age times at lower levels 
it is clear that summer temperatures as a whole were at best but little 
higher than now. We have however seen (p. 184) that if in our up- 
lands summer weather is predominantly anticyclonic the mean tem- 
perature deviates above the normal to a greater extent than in the 
lowlands. For example if in a dry sunny summer month the mean in 
the lowlands is, say, 2 above normal, that on the high Pcmiiws may 
be 3 above normal. Such an excess of temperature relative to the 
plains below would be just enough to raise the climatic tree line by 
five hundred feet or so compared with the present. Without doubt 
the balance of the evidence favours the view that Bronze Age summers 
were over long periods considerably drier, calmer, and probably 
rather warmer than at the present day, particularly in Ireland, 
Northern England and Scotland. In die south of England, at least 
near the Channel, the changes were less marked. It is interesting to 
observe how in the fine dry August of 1947, the lack of rainfall was much 
more marked in N.W. Scotland than in the extreme south. 

Evidence is as yet somewhat uncertain with regard to the happen- 
ings in Britain towards the end of the Bronze Age, say between 1,000 
and 500 B.C. Following minor fluctuations, at the end of this period it 
is evident that especially towards the north and west the climate again 
became considerably wetter, with more wind and cooler summers, and 
greatly diminished evaporation. There is, literally, overwhelming 
testimony in the massive mantle of peat which now covers all our high 
.flimmit plateaux unless they arc for one reason or another especially 
permeable or well-drained. Everywhere in our wetter districts more or 
less peat can be found; and in the Irish plain it becomes particularly 
prominent in the ill-drained lowlands. It can be seen at its grimmest 


and most intractable on Rannoch Moor or, farther south, covering 
smaller areas on Fairsnape Fell and the Kinderscout plateau on the 
gritstone Pennines. The balance of opinion favours the view that 
practically the whole of our hill-peat developed in a few centuries 
about 500 B.C. In many places it is now being eroded probably by the 
action of the weather, more quickly than it is being renewed by the 
further decay of vegetation; but our upland plant-cover is a complex 
matter on which reference should be made to Professor Pearsall's 
Mountains and Moorlands in this series. 

With regard to the 'Sub-Atlantic' the weight of the evidence 
suggests that in England the mean temperature of a summer month 
such as July was about 4°F. cooler than in the preceding optimum 
periods. This is not greater than the difference between a dull cool 
July nowadays, and a fine month. Compare July 1888 or 1922 (56 
in the N. Midlands) with July 1887 or 1921 (62°-63°). 

Exactly what caused this effective increase of wetness is still an 
unsolved problem. That it had many repercussions with regard to the 
movement of peoples is probable. Some have gone so far as to suggest 
that the great influx of northern peoples towards the Mediterranean, 
shown in the early invasions of Greece, can be ascribed to climatic 
causes. This is very uncertain; many factors besides climate have 
influenced Greek politics throughout all ages. In North- West Europe 
it used to be supposed that the invasions of Britain of the early Iron Age 
could be attributed to the more severe climate and particularly the 
colder winters about this time. Unfortunately more recent evidence 
of the spread of the beech supports the view that the winters were if 
anything a little milder than at present ; but the summers were definitely- 
cooler and more cloudy. The greater wetness and stormincss is plau- 
sibly reflected in a marked decline in the energy and vigour of 
Scandinavia as shown by archaeological finds. There is however some 
danger in drawing too many inferences with regard to climate from 
the cultural characteristics of peoples. It will be interesting to know if 
our descendants in two thousand years' time will be persuaded that 
-he arrival of an Iron Age in New Zealand about 1830 was due to 
starving refugees fleeing from the fearful winters of Napoleon's day. 

Opinion in recent years has tended towards the view that for some 
reason as yet unknown, which may for example be connected with 
minor variations in the intensity of the sun's radiation reaching the 
earth, operating over periods of lime which may amount to centuries, 



the vigour of the circulation of the earth's atmosphere varies con- 
siderably, at least around the North Atlantic and possibly more widely. 
If the circulation is feeble, the great cold anticyclones over the 
continents and the polar regions in winter build up more strongly and 
arc more persistent. Less frequendy than usual are they whittled away 
by the surging streams of air on their flanks or over the top, whose 
energy depends at least in part on those differences of temperature over 
the earth's surface which in turn ultimately derive from the sun 
through the variable medium of the atmosphere. Correspondingly in 
summer the great anticyclones over the oceans will tend to spread their 
influence over a larger area, and farther towards the Poles. At the 
present time we can observe such tendencies at work in individual 
seasons. In February, 1947, extremely persistent high pressure 
developed over the lands and seas of the Arctic. In the droughty hot 
summer of 192 1 high pressure extended with extraordinary obstinacy 
from the Azores to the Baltic; this was in part repeated in 1949 and 
1959. Possibly such tendencies developed over longer periods of years, 
arising out of slight variations in the general vigour with which the 
surface circulation was maintained; the dry quiet summer warmth of 
the Bronze Age might thus be explained. Earlier still, the amelioration 
of temperatures indicated by the spread of tree-birches during the 
"Allered oscillation" is again of the same order of magnitude as that 
which we find to-day between quiet anticyclonic summers and cool 
windy summers when the atmospheric circulation is more lively 
{Geographical Journal, March 1951). 

But referring for the moment to the Northern Hemisphere a second 
factor in the argument arises from the great variations in the extent 
of die Arctic sea ice. If a period of feeble circulation happens to 
coincide with a very icy Arctic we might expect a group of very cold 
winters; but if there was very little ice, the consequence might be a 
'Bronze Age' of dry warmth in the British Isles. For it is tempting to 
deduce from the evidence forthcoming from the North Pacific that in 
a region where the surface supply of cold Arctic water is impeded, due 
to the narrow Bchring Strait, the oceanic anticyclone takes up a much 

Plate XIXo: Chalk country: cumulus cloud over the Sussex Downs. Compare 

also Plate IVo, p. 47. 

b: Snow-laden winter sky over the High Chilterns, January 1936, near 
Ivinghoe Beacon. Trees show (he cfTcci of prevailing westerly wind. 


more northerly position in summer compared with the corresponding 
Azores High in the Atlantic. One result is that southern Vancouver 
Island now enjoys English temperatures in summer but with less 
rainfall and with 25% more sunshine. 

It certainly appears that the vigour of the circulation, represented 
by the frequency with which the contending air-masses surged over the 
British Isles and North- West Europe, was much increased five cen- 
turies or so B.C. That the increased circulation of the Sub-Atlantic 
followed a phase of increased spread of the Arctic Ice and cooling of 
the N. Atlantic surface waters is a hypothesis awaiting further results 
of ocean research (Ovey, p. 250). 

In Roman times the evidence favours die view that the climate of 
Southern England was still rather damper than at present; but it is 
extremely difficult to say how much of this was attributable to the 
extensive damp uncleared and undrained forest and march. Litde of 
the country was yet cultivated, although clearance of forest had 
probably begun. We obtain from Tacitus and other writers a con- 
temporary indication of temperature conditions litde different from 
the present. The frosts of winter were in general neither so severe nor 
so prolonged as on the adjacent continent, and no mention is made of 
frozen rivers or harbours. Attempts to cultivate the vine met with much 
trouble and little success. As for Tacitus's mists and fogs, many a 
peninsular Italian would say very much the same thing to-day; and it 
may be added that Prussians arc prone to remark on the rarity with 
which the cloudless blue summer skies of Pomcrania are to be seen, 
even in East Anglia. With regard to the effect of lack of drainage, 
Brunt has pointed out that over heavy soils the retention of water close 
to the surface will favour a decreased diurnal variation of temperature. 
But the evaporation of water from the ground surface would produce 
a superimposed lowering of air temperature by day and night. Forests 
would tend to check the development over wide areas of very low 
winter minima, but again the warming of the ground in spring and 
summer would be hindered. Nevertheless it must be remembered 
that there was undoubtedly a good deal of cleared land and open 
heath even in Anglo-Saxon times, on which conditions might be much 
as at present. 

In the Dark Ages following the Anglo-Saxon invasions there are 
few written records and the climate is as difficult to elucidate as any 
other aspect of that period. The weight of the evidence however 

H. Jenkins, Ltd. 
Plate XX: Trawlers going out at sunrise oil' the Norfolk coast, showing 
N.W. wind in late September with characteristic low cloud, breaking a 

little over the sea. 



favours the view thai the period centred about the sixth-scvciidi cen- 
tury A.D. was again considerably drier and less stormy in N.W. 
Europe. The amelioration of the Scandinavian climate led to better 
and safer harvests and to an enterprising, vigorous and energetic 
population, increasing rapidly enough for many to be willing to seek 
iheir fuiure overseas, and for others to display the splendid craftsman- 
ship and incipieni literary ability of which we have abundant evidence. 
Deplorable though this hearty accomplishment may have appeared 
to those who preferred the reminiscent ruins of the Mediterranean, it 
lies behind much of the present-day Britain. Many will reflect not 
merely on the natural landscape but also on the works of man. Lindis- 
rarnc and Whitby, the ancient trading streets of York and the fair- 
skinned matrons of quick-dealing pugnacious Leicester owe much to 
the spring-time north-easter round die Scandinavian anticyclone 
of twelve hundred years ago. Even the Cornish saints of the 
seventh century show signs of having profitably occupied the 
more permanent sources of water, in a period when the dry east 
wind blew more often than now. And in an Icelandic saga 
(Laxdxla) we read how Unn had ships built in Caithness where it 
would now be very difficult to find any trees well enough grown for 
the purpose. Pre-Conquest vineyards existed in England. 

On rather scanty evidence it seems that the climate tended to be- 
come wetter soon after the Norman Conquest. Much of our evidence 
depends on entries in monastic chronicles; and an entry from a single 
abbey in Kent of "a bitter winter with deep snow" might not in any 
way represent 1 he experience of a monk belonging to a less ascedc 
order in Devonshire. In the cold February of 1942, for example, the 
impression of wintrincss was very different in the Plynlimon moor- 
lands where there was but little snow, from that in Cambridgeshire. 
Careful investigations of all available medieval material have also 
been compiled elsewhere in Europe. To mention two; one inquiry 
concerns the frequency with which the Danish entrances to the Baltic 
were blocked by ice, while others have studied the frequency with 
which rivers such as die Seine at Paris were frozen, or gave rise to 
excessive floods at unusual seasons. The dates of harvest and of the 
vintage have also been studied. 

Further, reports of the success or failure of the harvest from S.E. 
England would not necessarily reflect the course of events in, say 
Galloway. A comparative investigation of the annual rainfall at 



southern English stations with that in the mountainous West High- 
lands has shown that for the last fifty years or so there has been a 
decided though not uniform tendency for a wet year at Oxford to be 
normal or even rather dry in the Argyllshire mountains and vice 
versa, that is, there is no significant correlation between the fluctua- 
tions in these two regions. Reference should be here made by those 
interested to papers by Dr. Glasspoole in British Rainfall, 1925. No 
year showed this more remarkably than 1937, which ranked as a very 
wet year indeed in parts of Sussex (over 150% of normal) while as the 
maps in British Rainfall (1937) show, at Ullapool on the north-west 
coast of Scotland the year was phenomenally dry, with only 61% of 
normal. A similar distribution prevailed in 1958. We have elsewhere 
seen that for many places an annual rainfall of about 65% of normal 
is likely to represent the "driest year on record" over periods up to a 
century or so in length. In general therefore wc must exercise much 
care in the interpretation of historic records of success or failure of 
crops. Further, with regard to the connection between the yield of 
crops and weather, temperature, rainfall and sunshine all play their 
part; a complex problem attacked by Hooker. 

All manner of fragments of information have been pieced together. 
A study of the distribution of medieval water-mills in Kent indicates 
that for a considerable period about 1275 the flow of the streams was 
deficient. It is extremely interesting to observe that about this time 
catastrophic droughts beset the western United States in the dust- 
bowl region with which the events of the past decade have made us 
fa miliar. Later, the frequency of reports of climatic catastrophes of one 
sort of another becomes so marked as to suggest that the fourteenth 
and early fifteenth centuries were a period of great variability. For 
example, we hear of great dearth in 1315-16. It has recently been 
shown that great inundations arising from tidal surges in the Southern 
North Sea arc associated with strong north-westerly winds accompany- 
ing the arrival of particularly deep depressions on the west coast of 
Norway. One of the most recent of such surges gave rise to the Chelsea 
floods in London in 1928. Many of the severe inundations of the later 
medieval period may have been associated with exceptionally deep 
Atlantic depressions of the same type. There also befell an extra- 
ordinarily severe winter in 1323, and a tremendous gale in January 1362 
— this last may have been the greatest gale ever known in Southern 
England, save the 'Great Storm' of November 1703. Our first British 


weather record, indicating the frequency of wind and rain at Oxford 
between 1337-44, nevertheless points to climatic conditions very like 
the present, perhaps with slightly milder winters. But the evidence from 
Scandinavia and Iceland points to a decline towards less favourable 
conditions from about 1300. It is possible that some great gales have 
left their mark on the scenery to this day. The great breach of the 
dykes along the Cardigan Bay coastal marsh land may have taken 
place in the sixth century. Fourteenth-century inundations are 
recorded from the Sol way as well as the east coast; great damage was 
done by inundations and blowing sand in Lancaslure in the sixteenth 
century and round the Bristol Channel in 1607. Some of these events 
and their effects are noted by Professor Steers (1946) ; and we may note 
the possible effects at Dungeness on the building of shingle-ridges. 

Later in the fifteenth century there is a little evidence of a return 
towards quieter and warmer conditions; among other things, the 
cultivation of the cherry spread northward, and it seems to have been 
known in county Durham in the sixteenth century, even at 800 feet. 
There is too, a long break in the reports of the freezing of the Thames 
at London — from 1434 to 1540. In Elizabethan days however, the 
climate as a whole appears to have turned a little cooler. Some 
mention should be made of the freezing of the Thames as no British 
river is better documented. The last year in which it was definitely 
passable on foot within London was 181 4. But in 1820 old London 
Bridge was pulled down and later the Lambeth marshes were em- 
banked; and as a consequence die tidal scour increased. There is 
little doubt that in any case with the same degree of cold the river is 
less liable to freeze over; the effect of old London Bridge for example 
was to jam the ice floes coming down the river. This is known to have 
occurred in 1740 and 1814. (Sec PI. XXII, p. 271). Hence caution is 
necessary in drawing conclusions from these old accounts. In 1895 the 
river was so nearly unnavigable that without doubt it would have 
become completely covered under the earlier circumstances. 

Mention of American data in a preceding paragraph will serve to 
remind many that the vicissitudes of annual rainfall, especially in the 
arid S. Western States, have been closely correlated with the successive 
growth-rings of trees. A chronology of the great droughts has been 
constructed from the evidence not merely of trees of exceptional age, 
but also from the beams used in the constructions of pre-Columbian 
buildings. Efforts have been made to extend this in more temperate 


climates bodi here and in America but die rate of growth of trees is 
influenced by too many factors to justify some of the earlier conclusions. 

Explanations of die entire course of European historic evolution 
from the indications of drought or additional rainfall given by a very 
small number of Californian trees have long been condemned as too 
bold. Nothing in the diversity of Western Europe operating on a 
variety of peoples through five thousand years of history, lends itself 
very happily to the solutions appropriate to a continent of greater 
simplicity both in space and time; not even the weather. Yet with 
the slow assemblage of data from various parts of both hemispheres it 
appears that whereas some climatic fluctuations may appertain largely 
to the Adantic margins others have affected the whole Northern 
Hemisphere and even die whole earth. For example the present 
retreat of glaciers is widespread. But with regard to climatic fluctua- 
tions we are still largely engaged in collecting and assessing the 
observational material. Some time must elapse before we can decide 
how far they are to be attributed to one or other of the several causes 
which have been brought forward in the past, such as changes in the 
amount of solar radiadou reaching the earth's surface, or in the extent 
and elevation of the earth's land masses. 

Careful studies have been made in recent years of the recorded 
movements of Alpine and Scandinavian glaciers. From these, the 
somewhat surprising conclusion has been reached that, since die final 
retreat of the last great glaciation, the greatest advance of the mountain 
glaciers of the Alps, in Norway and in Iceland has taken place within 
the past 400 years. The advance and retreat of glaciers has been 
shown to be influenced in the first place by variations of temperature. 
At intermediate levels in Norway, for example, a slight overall fall of 
temperature means that a greater proportion of the precipitation falls 
as snow rather than rain, and through the summer months there is a 
shorter period during which melting can occur to offset the greater 
accumulation of the past winter. A slight overall rise of temperature 
is best effected in a country such as Norway by an increase in the 
frequency of south-westerly winds from the warm Atlantic. This in 
turn commonly leads to increased precipitation, but often in the form 
of rain rather than snow; further for a great part of the year rapid 
melting is associated with the onset of the moist air, so that in 
the aggregate the glaciers retreat. Dr. Mannerfelt of Stockholm 
considers that the great retreat of the Ice Age was conditioned above 

C.E.S. R 



all by a greater frequency of warmer winds from the open seas to 
the south-west. 

In South-East Iceland, where the Vatnajokull ice-cap has been 
extensively studied, it has been shown that although over the last two 
or three decades the annual precipitation has increased by about 20%, 
the rise in the mean annual temperature of about 2°F. has been suf- 
ficient to lead to rapid retreat of the glaciers debouching from the 
ice-cap above. This ice-cap, the size of Yorkshire, lies close to an open 
*ea; many of our conclusions with regard to the character of Ice Age 
weather in Britain are derived from knowledge of what happens on 
these smaller ice-caps farther north. Recent investigations support the 
view that many if not all the smaller glaciers in Iceland and Norway 
arc not to be regarded as remnants from the Ice Age; they may well 
be the result of the deterioration which set in about 500 B.C. The 
semi-permanent Scottish snowbeds — our nearest approach to glacia- 
tion — appear to have responded in similar fashion. 

The greatest advances of the Icelandic and Scandinavian glaciers 
are now dated; taking the overall view, it appears that they reached 
their maxima about 1 745-1750; a second maximum was reached aboul 
1850. Since 1890 retreat has predominated, but it has not been 
regular; there was a slight tendency to advance about 1920, since when 
the retreat has become very marked. 

The course of events in Britain since Elizabeth's day can be roughly 
sketched. The material is surprisingly fragmentary, but it appears that 
her reign was characterised by a number of severe winters; the coolness 
or otherwise of the spring months remains to be examined. In 1564, 
the Thames was frozen; in January 1570 after the 'Rising in the 
North' had petered out the rebels were last heard of in the impassable- 
snowbound Cheviots. In 1571 and again in 1574-5 Derwcntwater 
was frozen, and the newly arrived Tyrolese miners had a lot of trouble 
with iced-up machinery. Stow mentions the extremely heavy snow- 
fall of February 1579 in London, and ihcrc was another heavy fall 
late in April of the same year. The spring of 1587 was very cold; 
another very cold April befell in 1595. Summers were variable, and 
for example, 1594 was excessively wet; but the .Armada summer of 
1588 was so magnificently English in all respects that it cannot be 
wondered that the nation was thankful for its weather. 

Many hints as to the weather come from the accounts of voyages. 
The cheerful departure from RothcrhiUie of the Muscovy Company's 


first trading expedition on 10 May 1553 gives us one of the most 
delightful pictures from Hakluyt; but as nearly three weeks then 
elapsed before they left Harwich we may well presume that in that 
year the Scandinavian anticyclone with its easterly winds was not 
only late, but very persistent. 

John Davis's quick passage to Greenland in June 1585 also hints at 
favourable winds; but as the scope of this book is confined to Britain 
we cannot go into the fascinating story of the Elizabethan seamen m 
the Arctic ice save to say that it was found in much the same place 
by the Victorians. 



WOO 1500 





Fio. 64 

Fluctuations in the mean winter temperature in N. Europe since 1250. Decline 

set in and prevailed about i550-t6bo; short, but marked amelioration about 

1720: minimum early 19th century (based on Easlon and Wagner) 

A better testimony to the general character of the late sixteenth 
century comes rather from Denmark, when Tycho Brahe's careful 
observations included the direction of wind. The frequency in winter 
of winds from points east of south indicates that in 1582- 1597 cold 
continental air then spread over Denmark more often than at present. 
His collected material has been considered to indicate a mean tem- 
perature for February and March about 2° or 3°F. colder than that 
of our recent decades, but these figures must be treated with a good 
deal of reserve. Nevertheless we may presume that a parallel effect, 
although probably less marked, would be found in England not dis- 
similar to that during the Napoleonic era, when a similar tendency 
prevailed, possibly a shade more extreme. Between 1579 and 1607 we 
have no reports of die Thames at London being completely frozen, 
as it was in 1795 and 1814. From the Alps there is evidence that the 
glaciers advanced considerably around 1600. It has even been sug- 
gested that the profusion of thick clothing characteristic of this era in 



England and Holland owed much to the climatic conditions; but 
something must surely be allowed for a demonstrative age in which 
the technique of heating dwellings was still very imperfect. 

Throughout the seventeenth century it appears that the fluctua- 
tions of climate were very similar to those we have experienced in our 
own fives. There were intermittent severe winters (1607, 1616, 1632, 
1658, 1676, 1683-4, ' 694-5, ar c noteworthy) and the concensus of 
opinion favours the view that following a poor wet year in 1648, 
Cromwell's Protectorate was favoured by several fine dry summers. 
That of 1652 was particularly fine in Scotland. Pepys' diary, and 
the many others of the period point to very much the same 
amount of outdoor enjoyment as we nowadays expect. For a 
detailed investigation of the material for this century the reader 
may be referred to a paper by J. N. L. Baker of the School of 
Geography at Oxford (1932). Indirect evidence comes from the 
attempts of the seventeenth century gardening enthusiasts to acclima- 
tise newly arrived plants that the climate differed little from the 
present. Nell Gwynne was a greater success as a saleswoman of 
oranges than John Evelyn as a producer; this feature arising from 
our climate has never been overlooked by Londoners, from Coven t 
Garden wholesalers to barrow-boys. We may continue to expect 
every effort to be made to persuade us that Italian peaches are to be 
set above Scottish raspberries. 

At last we are coming to the period of instrumental recording 
already mentioned in the second chapter, and here we shall interpose 
diagrams to illustrate the trend of temperature and rainfall during the 
past three centuries. To these we may add the results of interpretation 
of other early data. It must be remembered nevertheless that beibrc 
about 1800 all early instrumental records must be treated with (.he 
utmost care. The variety of errors to which meteorological obser- 
vations are liable, when made with doubtful instruments by pioneer 
enthusiasts in questionable exposures is tcrrifyingly large. 

From these diagrams and those for rainfall on pp. 266-268 the nature 
of the variations to which our climate has been subject will emerge. 
If with regard to temperature for example we smooth the curves by 
taking ten-year 'running means', we can get a better picture of the 
overall trend. It becomes evident at once that the temperature of the 
mid-winter months tended to be lower than at present over most of 
the early nineteenth century, whereas, spring, summer and autumn 


Ct'i'Ml Cduuo M40-MO 















•C i 1 = s 

t ( l 1} : : 1 1 1 , [ 11 1 • 1 ill I ] 
6tCAX CJ.S*9 



Fio. 65 

Ten-year running averages of the morality mean temperature for 

"Central England" from 1680-1960. These are broadly representative 

of the West Midlands; and before 1730 arc much less certain as they 

arc extrapolated from a distance. 


months show on the whole little difference from the average experience 
of recent decades. At the same time, however, it appears that 'severe 
winters' (characterised by a mean temperature below say 34 , and 
therefore marked by a predominance of snow instead of rain, together 
with a good deal of frost and probably considerable skating on still 
waters) tend to occur in irregular groups. Using this criterion we shall 
find a number of severe months between 1 688- 1 702, 1 740-48, 1 760-68, 
1776-89, 1795-1803, 1808-20, 1826-30, 1837-55, 1878-97, 1940-47. 
The trends apparent in the English and Scottish records arc 
supported by those in Holland, Denmark, Norway and Sweden. 


bdi tru ira on m mi mi im m m\ aai an si iss isu bsi 1*1 wri wn »ji koi mi un imi wi 
_ mo ins n.v iwo mo ireo itto irao ow> wo lao i$jd ojo umo bm isw bw mso ia» boo t»io tuo ism t*> i?» 
c # 1 • • 1 • • • 1 1 • • 1 1 1 1 1 1 1 1 1 

* — n — 

1 1 


'.1*1*41 in 

l'«Un4 ' 



i 1.1. ten 

u ..1715 


V .-1 IV, ,.,!!, 

t<ct««rn I7M - 1790 
ir»-ins k« «»,w 

Minr GljriA «lw*x of 
(radivuilvMi arttinoUocn 

p-J i|(kuI *<ln*Jr 

ci'-xmio*. before 1793 

1740- 174S 
0*T*w« 17*0) 

Cfflrrjlty Gmfral!? Strmo C..t-lrr.iM# MjstxJ Srtrul c*Htnun«. 

I»jiu.iu» UJ-Hli |MM SI19M AJ»»*e IM0-I9CO am »V> 

imu.,.11 or MO I* >=*K iittfwo * Stijhl «urur -J-»*l 

17*0- ISU kpMl MM ' tlKK* mO. lilt 

F10. 66 

Ten-year running means of spring temperature (March-May) in N. Eng- 
land and Holland (sec also Fig. 73, p. 281) 

Further, we must take note of the behaviour of the glaciers. There 
is evidence that the tendency towards amelioration of the winters 
between 1898 and 1939 was accompanied on the whole by a tendency 
to increased cloud and precipitation; for example, between 191 1 and 
1930 average precipitation at many British stations was about 5% 
higher than over the period 1881-1910 and the sunshine duration was 
slightly decreased. Since 1930 there has been an appreciable rise in 
the average temperature of spring, summer and autumn; but it would 
be unwise to assume a continuance. 


Studies of the frequency of wind from various directions have shown 
that the 'resultant' in N. W. Europe lies for 1901-1930 slightly more to 
the westward of south than formerly. In Scandinavia these facts are 
interpreted as the accompaniment of a renewed 'rise in the vigour of 
the circulation' — more frequent, and probably deeper depressions 
finding their way along the immemorial track from south of Iceland 
into the Norwegian Sea and onward towards Spitsbergen or, some- 
times, the Baltic; leading in turn to more frequent sweeps of the 
Atlantic air over that region, especially effective in raising the tem- 
perature in winter. These no doubt would affect Britain, especially 
in Scotland; we might appropriately say, the climate became more 
maritime during at least the three earlier decades of this century with 
a distinct reversal since 1930. In this period was a minor fluctuation 
represented by a slight advance of the Norwegian glaciers, within the 
general retreat, and in England by the rather cold winters and springs 
of 1917, 1919, and 1923 and the cool summers of 1919, 1920 and 1922. 
But the lack of consistency even in a brief spell of this kind is shown by 
the fact that in Southern England 1921 was particularly warm and dry. 
While it appears that fluctuations of some magnitude develop, yet 
as far as we can see they arc not necessarily periodic, continuous as 
regards any given characteristic or of uniform length. The picture is 
only slowly emerging. If in the decades when Stonehcnge was new there 
were eight fine summers and but two were wet, we recall within our own 
experience the general opinion that in the wet 1920's and 1950's there 
were two very fine summers and eight wet. Yet rarely have the 
extremes been so great that there was even widespread failure of crops. 
Over four thousand years the amplitude of variation has not been large. 
One of the most significant fluctuations which can just be recog- 
nised in the instrumental period has been mentioned, namely, the 
decided swing towards greater dryness, in Southern England at least, 
through the earlier part of the eighteenth century. Continuous series 
of rainfall measurements are available for Lancashire (1677-1704) and 
in Essex (1697-1716) but then there is a gap until 1727. From 1727 
the overlap of each record can be compared with others in a con- 
tinuous succession down to the present day. This is necessary in order 
to eliminate several possibilities of error in measurement. Although 
for many years there were but two gauges, they were so placed as to 
enable us to judge with considerable accuracy the extent of the 
variations over England as a whole. For Scotland however, we cannot 


begin a series until 1785, by which time quite a number of records were 
becoming available farther south. 

All the indications show that the period 1700- 1750, possibly 1680- 
1 75°) gave at least in southern and eastern England averages of rainfall 
distinctly below the 1881-1915 normal. Since 1727 the driest decades 
in England have been 1740-49 (86%) and 1850-59 (93%); compare 
die diagram on page 266, based on Dr. Glasspoole's deductions. Four 
years in succession, 1740-43 were phenomenally dry; fortunately, the 
rainfall occurred just sufficiently and at the right time for the grain 
crops. In the interval for which measurements are lacking contem- 
porary writers name 1716-19 as very dry; and in S.E. England 1714 
seems to have been one of the driest years ever known, comparable 
with 1788 and 1921. 

It is very tempting to relate all this to the English scene. How 
much do we owe to the great agricultural improvements of that age 
when harvest after harvest served to encourage the enterprising 
farmer? The era of comely brick in the Norfolk market towns; 
the rising revenues of College estates; the speculative fever of the 
South Sea Bubble may owe more to climate than we think. Farther 
north, the widespread rebuilding of Lake District farmsteads from 
1680 onward has drawn attention. Domestic buildings in stone had 
been erected for at least a hundred and fifty years; but all over the 
wetter Pennine dales signs of a quiet prosperity are to be found which 
together with miles of stone wall indicate that there was energy to 
spare. In the rise and spread of Quakerism; in the establishment of 
country banks; in the spare energy available leading to the widespread 
cultivation of remarkable observational talents by North and West 
country medical men climate may have played its part. 

The drainage of fields and clearing of woodland no doubt assisted 
the generation of farmers who had to face the wetter years of the 1 760's 
and early 1 770's, the cold seasons of 1 782, 1 784, and 1 799. Rising prices 
too made it here and there profitable to attempt the reclamation 
of land far up the hill sides in the Napoleonic Wars, in spite of the 
setbacks which must have beset the upland farmer in cold summers 
such as that of 1816. Eventually, however, the conflict of agricultural 
and industrial interests was resolved; the years from 1820 to 1845 
were more often wet than dry and the succession of poor harvest and 
dearth did much to foster the repeal of the Corn Laws. In the north 
the rising industrial towns were largely recruited from the country 


population. The immense profusion of present-day surnames o£ 
territorial origin in Lancashire and Yorkshire industry cannot fail to 
be noticed by those who contemplate the names of the fellside farms on 
die one-inch Ordnance Survey. 

The fluctuations of early industry were not unaffected by the 
vicissitudes of die wcatiier. Shortage of water for mill-lodges and canals 
during the intensely hot dry summer of 1826 played a big part in the 
distress of that year, partially relieved by widespread road-building in 
Lancashire. As a further consequence we may note a considerable 
increase in the number of rain gauges kept on either side of the 
Pennines; incidentally, the predominance of dry summers in the 1850's 
led to a further interest in rainfall measurement and was followed by 
the establishment of the British Rainfall Organisation in i860. Our 
earlier industrialists were dependent on weather in many ways; in 
1838 a prolonged east-wind spell led to much delay and difficulty for 
the ships bringing die cargoes of cotton into the Mersey. 

The diagrams will serve to illustrate the extent to which tem- 
perature has varied. Apart from the wide swings shown by the 
predominance of colder winters in the Napoleonic, the early Victorian, 
and the late Victorian eras, occasional exceptionally cold summers 
owe their character, dismal from the farmers' point of view, to volcanic 
eruptions in distant lands. Such were 1784, 1816, possibly 1845, i860 
and 1885; also 1902 and 1912. Preceding each of these years a violent 
explosive eruption of ash, characterised by the spread at high levels 
of minutely fine dust, resulted in a large part of the earth's surface 
being to some extent screened. It is in summer, rather than winter, 
that the effects of such screening of the incoming radiation become 
most prominent with regard to temperature. 

After premonitory rumblings, the great Icelandic eruption of 
Laki (Skaptarjokull) early in June 1783 led not only to prolonged 
darkening of the sky but to a rain of ash which damaged the pastures ; 
the series of catastrophes following this eruption was perhaps the worst 
in Icelandic history. About two weeks later observers in England, 
among them Gilbert White 1 and Parson Woodforde, began to 

1 Gilbert White's version is: "By my journal I find that I had noticed this 
strange occurrence from June 23rd to July 20th ( 1 783) inclusive, during which period 
the wind varied to every quarter without any alteration in the air. The sun, at noon, 
looked as black as a clouded moon, and shed a rust-coloured ferruginous light on 
the ground, and floors of rooms; but was particularly lurid and blood-coloured at 
rising and setting." Letter LXV to the Hon. Daines Barrington. 


comment on the hazy sky, so marked that "the sun appeared red until it 
was twenty degrees above die horizon". Ash also fell in Norway and 
the north of Scotland. July was extremely hot and sultry, but of the 
next twenty-one months in the north of England all but three were 
colder than normal; the effects of the cold summer of 1784 were very 
evident in Scotland, although slightly redeemed after September 
which appears to have been the only month with a mean temperature 
above normal during that chilly year. The chilly year 1816 followed 
a scries of violent East Indian eruptions. The most recent, 1912, was 
marked by the coldest August on record at many English stations; 
summer throughout was also exceptionally cloudy, and when it was 
clear many observers commented on the remarkably pale blue colour 
of the sky. A violent eruption of Katmai in Alaska, in April, appears 
to have been the cause. 

He who reflects upon any view of British countryside cannot but 
be struck by the extent to which it is man-made, except when one 
crosses the fiftecn-hundred-foot level or takes refuge in some carefully 
restricted remnant of marsh, heath or dune. Man's effect on the 
landscape has however been influenced throughout history by climatic 
factors. There have been times when nature and man have been in 
harmony. The age of reason may also be the age of reasonable weather. 
In other periods the crops have only been brought home with great 
difficulty or anxiety; many have been led through desperation to seek 
a living elsewhere if no better solution presented itself at home. There 
have been sheltered corners where a secure livelihood widi little effort 
could for long be counted on. Everywhere something of both clime 
and time can be seen in buildings. To this day the universal 
semi-detached villa still shows in the varied exotics of its garden 
the Englishman's conscious desire to improve and adorn his 
own bit of the island, while the poorly fitted windows and 
plumbing reveal the ever-present temptation to forget that extremes 
occur, begotten of die maritime climatic phase of the early 
twentieth century and the desire for quick returns. It was when 
the climate was most maritime between the wars that we were most 
tempted to disregard the continental menace. Who knows that 
there is not a link between our collective attitude of mind and the 
airs that blow upon us, bringing with them a reminder that 
somewhere beyond the wind are other men? 




Chap 12 
Ahlmann, H. W. (1948). Glaciological Research on the North Atlantic 

Coasts. R.G.S. Research Scries, 1. London. 

(1949). The Present Climatic Fluctuation. Geogr. J. 112: 165-93: 

gives many further references. 
Angstrom, A. (1939). The Change of the Temperature Climate in Present 

Time. Geogr. Annaler Stockholm, 21: 119-31. 

(1946). Sveriges Klimat. Stockholm. Gcneralstabcns Litograliska 

Baker, J. N. L. (1932). Climate of England in the Seventeenth Century. 

Q.. J. Roy, Mel. S. 58: 421-38. 
Brooks, C. E. P. (1930). The Climate of the first half of the Eightcendi 

Century. Q,. J. Roy, Mel. S. 56: 389-402. 

(1949). Climate through the Ages. London, Bcnn, 2nd cd. : the standard 

work, first publ. 1926. 

and Glasspoole, J. (1928). British Floods and Droughts. London, Bcnn. 

and Hunt, T. W. (1933). Variations of Wind Direction in the British 

Isles since 1341. Q,. J. Roy, Met. S. 30: 375-88. 
Brunt, D. (1945). Some Factors in Microclimatology. Q..J. Roy. Met. S. 7/: 

Charlesworth, J. K. (1957). The Quaternary Era. London, Arnold. 
Flohn, H. (1954). Witterung und Klima in Deutschland. Berlin. 
Glasspoole, J. (1928). Two Centuries of Rainfall. Mel. Mag. 63: 1. 

(1931). General Monthly Rainfall over England and Wales, 1727- 

193 1. British Rainfall, 71: 299. 

(1937). Rainfall over the British Isles, 1901-30. British Rainfall, 77: 

Godwin, H. (1941). Pollen Analysis and Quaternary Geology. Ptoc. 

Geol. Assoc. 52: 328-47. 

(1947). The Late Glacial Period. Science Progress, 55 :■ 185-92. 
Harrison, J. W. H. (1948). The Passing of the Ice Age. New Nat. J. t: 

Hooker, R. H. (1921). Forecasting the Crops from the Weather. 

Q.. J. Roy. Met. S. 47: 75-99. 

(1922). The Weather and the Crops in S. England, 1885-1921. 

Q_. J. Roy. Met. S. 48: 1 15-38. 
Labrijn, A. (1946). Climate of Holland in the past two and a half 

centuries. Med. Verh. K. Ned. Mel. Inst. Utrecht, 49: (No. 102). 
Lamb, H. H. (1959). Our changing climate, past and present. Weather 

14: 229-318. 



Manley, G. (1946). Temperature trend in Lancashire, 1753-1945. 

Q.. J. Roy. Met. S. 72: 1-31 : gives many references to older data. 

(1949). The Snowline in Britain. Geogr. Annalcr, 31: 179-97. 

(1915). The range of variation of the British climate. Gcogr. J. iij: 


('953)- The mean temperature of Central England. Q,. J. Roy, Met. 

S. 7g: 242-261. 

('O^)- The late-glacial climate of N.W. England. Liv. Manch. 

Gcol. Journ., s, 188-215. 
Meyer, G. M. (1927). Early water-mills . . . E. Kent. Q.J. Roy. Mel. S. 

53: 407-429. 
Ovey, C. D. (1950). On the interpretation of climatic variations . . . 

Atlantic deep-sea core. Centenary Proceedings, Roy. Met. S., 21 1-2 15. 
Pearsall, W. II. and Pennington, VV. (1947). Ecological History of the 

Lake District. J. Ecol. 34: 137. 
Fenningto.n, Winifred (1947). Lake Sediments of Windermere. Philos. 

Trans. Roy. Soc. London, 233B: 137-75. 
Schove, D. J. (1949). In discussion on post-glacial climatic change. 

0_.J.Roy.Met.S. 75 : 175-79. 
Simpson, Sir G. C. (1934). World Climate in the Quartcrnary. Q.. J. Roy. 

Met. S. 60: 425-78. Also ibid. 83, 459-80 (1957). 
Stamp, L. Dudley (1946). Britain's Structure and Scenery. London, Collins' 

New Naturalist. 
Steers, J. A. (1946). The Coastline of England and Wales. Cambridge, 

University Press. 
Stow, J. (1631, cd. Howes). Annales: or a general chronicle 0/ England . . . 

London, 685. 
Wheeler, R. E. Mortimer (1943). Maiden Castle. Research Series XII, 

Society of Antiquaries, London. 
White, Gilbert (1789). The Natural History and Antiquities ofSelborne . . . 

Wright, W. B. (1936). The Quaternary Ice Age. London, Macmillan. 
Zeuner, F. E. (1945). The Pleistocene Period. London (Ray Society). 

(1949). Dating tlie Pas'. London, Mclhucn, 2nd cd. 


The recent discovery of meteorological journals kept since the late 17th 
and early 181I1 century has made it possible to estimate the mean monthly 
temperatures characteristic of the West Midlands from 1680 onwards. 
From 1 680-1 730, however, estimates arc based on rather distant stations 
with very imperfect instruments, so that the early part of their answers 
should be viewed with caution. 



Fll show thee 

All the qualities o' the isle 

Shakespeare: Caliban, in the Tempest 

For the benefit of those whose curiosity is better satisfied by 
figures there follow some comments on the extremes shown in 
our statistical tables. The upper limit of possibility of overall warmth 
in January seems to have been approximately demonstrated in 1 796, 
1834 and 1 9 16. In each of those years mean temperature for Midland 
stations lay in the region of 45° (6° or more above normal) and the 
thermometer at many inland places scarcely sank to freezing-point, 
if at all. At a great many stations not even sleet fell. 

Such warm winter months are easily explained; provided that 
a continuous series of depressions passes on a track from on" the 
west to the north of Scotland we remain throughout in the milder 
type of maritime polar air with intervals of maritime tropical. 
Not only are the skies cloudy or overcast for long periods; the air 
has practically no opportunity of stagnating for more dian a very 
few hours. 

It is interesting to observe that if such conditions were to persist 
through the winter months the consequences would appear to resemble 
to a marked degree those of the Atlantic phase (cf. Chap. 12, p. 238) 
of warmth and moisture about 4,000-5,000 B.C. In summer, warm 
humid weather with a predominance of moist air from southerly points 
is represented by such a month as July 1852, or less markedly by July 
1928. The July of 1852 was one ol the hottest on record in Scotland 
and N. England; at the same time it was particularly humid with 


a good deal of rain. Mean temperatures at sea level were ol 
the order oi 64° in ihe wetter north-west; at London, nearer ihe 
continent and frequently in a drier and less cloudy air-mass, ihe mean 
exceeded G8 e . 

In other words if the continental and Arctic sources of cold air 
were absent or not effective, and yet the Atiantic air circulation 
remained lively, we might expect a mild moist climatic phase rather 
like the wetter parts of New Zealand, and probably rather more cloudy. 
There would be occasional slight frost; it would not be as warm as the 
Azores, but many delicate evergreen shrubs might be expected to 
flourish everywhere. 

Under such conditions the orographic component in the rainfall 
of the western parts of Britain would become even more marked. To 
illustrate this we may consider the data for March 1921 and November 
1938, both being south-westerly months almost throughout. 

March IQ21 Pressure 5 mb. above normal in Kent, 4 mb. below 

normal in Hebrides. 
Mean temperature 2 "-4° above normal everywhere, 

S.W. dominant. 
Rainfall (England S.E., E. and N.E.) 47% below normal. 

(England S.W., N.W. and Scodand W.) 31% 

above normal. 

November 1938 Pressure mainly normal in Kent, 11 mb. below normal 

in Hebrides. 
Mean temperature 4°-5° above normal everywhere, 

S.W. dominant. 
Rainfall (England S.E., E. and N.E.) 13% above normal. 

(England S.W., N.W. and Scodand, W.) 73% 

above normal. 

In each of these months, as the diagrams, data and maps in the 
Monthly Weather Report show, winds from between south and west 
predominated to such an extent that temperature was well above 
normal. It is not difficult to imagine our climate changing slightly in 
diis direction. If such a change were to occur with rather more 
emphasis over a longer period we should find that over a period of 
years the rainfall of all our western hill districts would be increased 
much more markedly than that of die lowlands farther to the south 


and cast. Evaporation too would diminish; and assuming that the 
country was no lunger artificially drained the result would be increased 
water-logging of the soil and eventually the formation of extensive 
blanket-bog and peat. This has been and remains characteristic of 
the wetter phases of the British climate. 

Now let us look at the worst. If continental-Arctic air prevailed 
in winter, gradually turning to maritime-polar and maritime-arctic 
in spring, summer, and autumn, the consequences would be revealed in 
a series of monthly averages, for lowland stations inland, somewhat 
as shown below. (For convenience the actual months based on avail- 
able long averages in which such extremes were recorded are also 
given.) It will be seen that the resultant would be very similar to the 
present climate of Iceland in winter, and a little warmer in summer. 
It might be still possible to grow oats or rye in the Midlands. Apples 
would be rather doubtful, and wheat almost impossible. Permanent 
snow beds would be found on Snowdon and in the Lake District; on 
Ben Nevis and in the Cairngorms there would after some years be 
small glaciers. There would still be abundant pasture and rather 
stunted trees at lower levels in England. 

Never in the past two hundred years have such conditions, either 
of warmth or of cold, persisted through a whole year. In the warm 
years 1834 and 1949 they were nearly approached. On the cold side, 
the wet, snowy and chilly year 1695 appears to have been one of the 
worst wc know of, while 1740 was dry and cold almost tluoughout. The 
period 1692-1701 was particularly unfortunate in Scotland, with its 
succession of cold late springs and inclement wet summers. In Scotland, 
the same combination of cold spring and wet cool summer rendered 
1782 exceptionally bad; in this respect 1799 too, was bad. Indeed it 
is generally true that cloudy and wet cold years are the worst of all, 
such as 1879 in southern England — the year which, coming after a 
scries of wcttish summers, was 'the ruin of English agriculture'. In 
Southern England 1958 and i960 have been wet, but not cold. 

Anticyclonic weather with its accompanying warmth and dryness, 
rather than warmth and humidity give south-east England the highest 
summer average temperatures. To illustrate the possibilities of dry 
warmth wc may name 1733, and the three successive hot summers of 
I779> I 7^° and 1781; subsequently 1826, 1846, 1868 and at least in 
the south 191 1, 1 92 1, 1933 and 1947 were outstanding. To these we 
may now add 1949, 1955 and 1959. 

254 climate and the british scene 

The Range of Mean Monthly Temperature: 

Warmest and Coldest Months on Record (1751 to date) °F. 

(Lancashire Plain) 




453 46-5 48-3 5<>-5 58-5 64-0 648 64-8 6 i-o 55-4 485 46-4 
1916 1779 1957 & 1833 1C46 1808 1947 1865 19591818 1934 

1852 1938 



25G 28-9 33-5 399 46-5 52-9 553 54-3 48-7 42-6 35-0 30-0 
1814 1855 1785 1837 1782 1823 1816 1912 1807 1817 1782 1878 


1 75 1-1050 

39-5 39-6 4>- s 455 5' 6 5^0 59-4 586 551 49-3 423 398 
39o 395 4 2- 5 4 G- 5 5'7 57° o0 3 597 559 49'8 43'3 4 0- o 

34-4 39-0 41-2 45-9 51-6 57-1 59-9 59-2 552 48-7 42-2 38-9 



♦Estimated. Next coldest, 1879. Back to 1680, none of the above records 
appears to have been seriously surpassed. 

Mean Temperatures for Each Month: 
Exceptional Years (Lancashire Plain) 

1 1 appears that once in a century or so the mean July temperature 
within London reaches 70 ; and between 69° and 70 at south-country 
stations in lowland valleys. 

So long as Britain is an island surrounded by unfrozen seas the 
possible range of climatic conditions for individual months lies very 
nearly within the extremes illustrated by the last two centuries. For 





440 42-1 44-4 45-8 548 58-6 625 605 55-7 50-8 44-1 420 

44-7 4'-' 44-5 4<>-» 5'5 574 63-0 58-5 56-5 55-3 40-5 43-6 
344 4°8 45' 4 8-2 553 5 8- 2 6r-6 61-9 583 554 451 427 



Warm and 
rather wet 
Warm and 
rather dry 


39 395 425 46-5 5>-7 57° 603 597 55-9 49-8 43-3 400 



34-8 364 376 408 484 575 585 573 550 46-5 409 34-3 
30-5 368 403 42-1 479 54-9 56-2 57-8 540 479 40- 1 33-3 


Cold and 

rather wet 

Cold; wet 



it is evident that with an unfrozen sea the lowest possible mean tem- 
perature in an island of our size can scarcely fall below about 24 at 
sea-level. This figure was nearly touched in January 18 14; reasonably 
well-exposed thermometers at Perth and Carlisle gave means just 
below 25°. 

Within this frame wc may now look at the incidence of extremes 
with regard to maxima. The highest outdoor temperature we are ever 
likely to get in the hottest parts of S. England lies very close to ioo° as 
in July 1808, 1868; August 191 1, 1932. It must be remembered that 
there is always some doubt attached to extremes before 1880 or so as 
they were not always kept under conditions we should nowadays 

In Scotland, 90 has been very rarely reported (Lcith 1876, 
Prestwick 1948) but so many stations in later years have had values 
of 87 to 89 (Paisley, Elgin, Kilmarnock, Liberton, Perth) that it 
seems likely that some part of Scotland should attain 90 perhaps 
once in twenty years, while in a city such as Manchester 95 might 
be attained. 

The lowest values ever likely to be reached in exceptional locations 
under severe conditions arc indicated by the records kept in certain 
well-defined frost-hollows when a deep fresh snow-cover was also 
present; and even in the London suburbs occasional minima below 
zero have been known, the first we hear of being in January 1684. 

Many places in frost-hollows from Scotland to Sussex have in 
recent years recorded between — 3 and —6°; but the further drop 
below — io° seems to be very rare indeed in S. England and rare in the 
Midlands and Scotland even in the most favourable sites; Braemar 
recorded — 13 in February 1955. Indeed unless frost-hollows arc 
chosen, temperatures below zero are rather rare and in the earlier 
nineteenth century grave doubts were cast on anyone who claimed that 
such readings had been obtained. In eighty years at Cambridge zero 
has been touched once and +1° twice (in February 1947); in a 
hundred years at Durham the minimum has fallen once only (1895) 
to i° below zero; while at Stonyhurst, up on a slope above the Ribblc 
valley, nothing below 4 has been observed since 1848. Cold air at 
this temperature may, however, spread over snow-covered open plains 
and down the broader valleys to the coast; Blackpool has recorded 
-i° (in 1881), Swansea +2 (in 1945), Southport +2° (in 1881), 
Barnstaple +3° (in 1945), Aberdeen -f-4 (in 1895). 
c.b.s. s 


It must be remembered that such extremes only affect a narrow 
surface layer of air in exceptionally exposed open valleys and plains. 
On upland ridges a mere three hundred feet above, quite normal 

minima may occur on the same 
night. At Little Rissington, 730 
feet up on the Cotswolds, the 
lowest minimum for any of the 
exceptionally severe nights of early 
March, 1947 was 21 , whereas in 
the Severn valley figures round 
zero were recorded. 

More remarkable, because 
more dangerous, are the very 
severe frosts which can occur in 
spring; and in the other direction, 
some very surprising high temper- 
atures have been known in quiet 
anticyclonic weather as early as 
the end of February. Maxima of 
75° near Harrow on 9 March 
1948 (map on page 88) and 77* 
at Wakefield on 28 March 1929 
were as outstanding as a mini- 
mum of 5 near Penrith on 2 
April 1919. Going back into early 
records, 67 was recorded at 
noon on the north wall of Gordon 
Castle as early as 10 March 1826. 
Without doubt this implies the 
possibility before the middle of 
March of a shade maximum over 
70 in many parts of the far 
north of Scotland. If such things 
can happen 70 will occasionally 
be reached in February in 

F10. 67 
Exceptional warmth with a south 
wind on the Moray Firth, 12 h., 
27 November 1948 (59° at Ix>ssie- 
mouth); dense fog and low temper- 
atures throughout S.E. and Midlands 
to N. E. England. Partial fohn in lee 
of mountains (Notation, p. 5) 

southern England; perhaps once 
in 200 years. In the autumn, Cambridge had a maximum of 70 on 
5 November 1938. As this exceeded by three degrees any previous 
November maximum anywhere in England since 1880, there is litde 


doubt that such warmth in November is probably almost as rare as a 
similar occurrence would be at the end of February. For the highest 
mid-winter temperatures we must look to the possibilities of subsiding 
air combined with fohn as we have already seen (p. 130). 

Some extremes for each month are summarised in the table below. 
Although some of the records are unofficial, a good general idea of the 
variability likely to be found at inland stations will be obtained. 

Low Temperatures: Recorded Extremes (°F.) 

Jan. * -20, Berwickshire 
-16, Kelso 

* -14, Walton-on- 


Feb. -17, Bracmar 

* -13, Bracmar 

* -20, Aberdeenshire 

Mar. -6, Houghall 

-9, Logic Coldstonc 

April 5, Penrith 

May 15, nr. Thctford 

18, Bracmar 

19, Appleby 

June 22, Dalwhinnie 

1 88 1 July 



'955 Aug. 

1855 S 

1947 Sept. 
J 958 

I9'7 Oct. 

1 89 1 Dec. * 
1955 * 

28, West Linton 

(Peebles) 1926 

29, Scodand, 

numerous 1919-58 

30, Santon Downham, 

Norfolk i960 

27, nr. Alston 

(Cumberland) 1885 

28, Glenlivct 1954 

20, Dalwhinnie 1942 

21, Braes of Glcnlivet, 

Banffshire 1948 

1 1 , Bracmar 1 880 

Dalwhinnie 1 948 

13, S. Farnborough 1926 

-10, Braemar 1919 

-20, Berwickshire 1879 

-15, Chcadle (Staffs.) 1860 

-11, Maidstone (Kent) 1796 
(♦approximate, reasonably authentic) 



High Temperatures: Some Extremes on Record (°F.) 
(Excluding London) 

63, Aber., 1929; Rhyl 1916 
62, Durham 1888 

61, Wrexham 1944 

67, Cambridge 
67, Barnstaple 


Mar. 75, Wcaldstone 1948 

77, Wakefield 1929 

74, Ilaydon Bridge 1957 

April 83, Cambridge 1893 

81, Peterborough 1945 



High Temperatures: Some Extremes on Record (°F.) 
(Excluding London) — continued 


May 91, Tunbridge Wells, 

etc. 1944 
89, Mildcnhall 

(Suffolk) 1947 

June 95, Rickmansworth 1947 

94, Waddington 

(Lines.) 1947 

95, Norlholt 1957 

July 96, Cambridge 191 1 

95, Reading 1923 
(>95) S.E.England 

1868, 1825, 1808, 1757 

94, Cromer, etc. 1959 

97, Halstcad (Essex) 1932 

96, Cambridge 191 1 

96, Norwich 1932 

Sept. 94, Raunds 

(Norlhants) 191 1 

Oct. 83, Reading, etc. 1921 

83, Rugby, etc. 1959 

Nov. 71, Prestatyn 1946 

70, Cambridge 1938 

Dec. 65, Achnashellach 

(N.W. Scotland) 1948 

The range of possibilities with regard to rainfall and snowfall can 
be briefly summarised. Rainfall statistics over long periods have been 
closely analysed; and over the country as a whole it appears that at 
most stations the driest year can be expected to have about 60% of 
the normal rainfall, the "wettest" 150% taken over periods of the 
order of a hundred years. With regard to the months, many places in 
the British Isles have recorded a whole month without measurable 
rain; even in the very wettest districts this has been known to occur 
in months with persistent wind from unusual quarter*. For when 
north and east winds prevail our normally wet highland districts lie 
on the lee side of the country. Hence in February 1947 no pre- 
cipitation whatever was measured at Glenquoich, in the wettest area 
of the Highlands; and in February 1932 the like occurred at the head 
ol Borrowdale. In August 1947 no rain whatever was recorded in an 
extensive area towards the west of Scotland. Months in which anti- 
cyclonic conditions have so dominated the country that a wide area 
recorded no rain include September 1865, February 1891, June 1925, 
February 1932, April 1938, September 1959. Many other spells, how- 
ever, have obviously occurred in which absolute drought prevailed for 
a period upwards of thirty clays though not coincident with calendar 
montlis. One of the most severe droughts on record lasted through the 
greater part of April-May 1893. Studies of the incidence of dry spells 


have been made which indicate as we should expect that the greatest 
tendency for such spells is in the spring and early summer months; 
there is a minor peak in September. The least chance is in the late 
autumn, but no month is entirely devoid of dry spells. The chance 
that a spring month such as May will have no rain can, however, be 
estimated as several times greater than a rainless August; while the 
phenomenal dryness of August 1947 is fresh in memory it is worth 
reiterating that dry Augusts are rare; August 1955 was notable. 

Snowfall is an immensely variable element as we have seen. If we 
regard the snowincss of a month as represented by frequency and 
aggregate depth of snow, probably the most snowy months in the past 
two centuries over the country as a whole were February 1947 and 
January 1814. Locally snowfalls have been heavier; for example the 
great Lancashire snowstorm of January 1040 and the phenomenal fall 
in Northumberland and Durham in February 1941. In the south, 
the Christmas snowstorm of 1927 gave remarkable depths in Kent, 
Sussex and Hampshire; the classical blizzards of December 1836 and 
January 1881 affected the whole of the south and Midlands, while 
that of March 1891 was very heavy in the west country and classical 
on Dartmoor. In Eastern Scotland where for orographic reasons 
heavy falls are fairly frequent, March 1886 was very exceptional. 
Farther back there have been some remarkably deep out-of-scason 
falls; in September 1673 and October 1836 in the county of Durham, 
in May 1782 and 1923 in Aberdeenshire, in June 1749 and 1809 in 
parts of Scotland, in May 1838 in the Cotswolds, in April 1908 and 
1919 in Suffolk. All through our annals miscellaneous accounts of 
heavy falls can be found; in January 1751 a prevailing depth of 27 
inches was noted at Richmond in Yorkshire, just as it was in February 
1933. In February 1567 and January 16 15 very great losses of sheep 
followed heavy snow in the same area. The same climatic factors 
have long been in operation, so that in many parts of Great Britain 
the results of a given combination of falling barometer, north-easterly 
wind and upland exposure are long remembered. 

Snow has fallen even on low ground in June; in the northern 
counties of England and Scotland it may perhaps be reported here 
and there as sleet once in thirty years. It is stated to have fallen on 
11 July 1888 at several places from Cumberland to Kent, although 
when we look up the meteorological events of the day in question it 
may be suggested that while the day was undoubtedly very chilly the 



distinction between sleet and soft hail was not at that time everywhere 
clearly drawn. Heavy falls of hail in summer have been claimed as 
snow in quite recent years, when in 1930 some keen Yorkshircmen 
ran charabanc excursions "to sec the snow in August" after a violent 
thunderstorm on the Yorkshire Wolds. Sleet has been reported in 
August from one or two high-lying Scottish stations, and, doubtfully, 
from Gordon Castle on 17 August 1784; also from Dartmoor in 1879. 
But Defoe's dating of a late August snowfall on the moors between Roch- 
dale and Halifax (in 1705?) must be doubted. Even on our lower 
mountains however snow has occasionally fallen to some depth in June. 
In July and August it is a little doubtful whether, since 1900 and south 
of the Border, a lasting cover below 3,000 feet has been observed at 
all. Generally speaking, if a keen watch is kept, Scottish summits above 
3,000 feet will be found to be freshly covered for a short time on about 
one occasion in June, and for the first time in autumn, during die 
latter half of September. 

Bearing in mind that the water equivalent of a foot of snow is 
generally taken to be an inch of rain, it is evident that for a single fall 
to exceed ten inches in depth is very rare on low ground inasmuch as 
falls exceeding an inch of rain in a day arc dicmselvcs very few unless 
in occasional severe thunderstorms. Depths on die level, apart from 
drifts, have been known to exceed four feet, for example in upland 
Denbighshire and Upper Tecsdalc in February-March 1947, die result 
of accumulation of several heavy falls in those districts in which as we 
have already seen snowfalls arc greater in amount on account of relief 
and easterly aspect. 

From County Durham after a single prolonged fall four feet was 
reported from Consctt (800 feet) in 1941, with 42 inches at Durham 
Observatory (336 feet); snow is said to have fallen condnuously for 
between 50 and 60 hours. The effects even of low hills are shown by 
die falls of February 1947 in the east Midlands. At the time when about 
nine inches was the prevailing depth in Cambridge, at diree hundred 
and fifty feet above sea level in west Bedfordshire depths of over twice 
this amount prevailed. Similar dcpdis were attained on the low 
Lincolnshire heights and on die rising ground in North Norfolk. 
Many northern motorists will remember how often the difference can 
be seen between the depth of snowfall at Cattcrick Bridge and that at 
Scotch Corner, five miles to die northward and 300 feet higher. It is 
here that the Glasgow-bound driver begins to wonder what is in store 


for him on Stainmore, that summit which his ancestors have known for 
so many centuries. Bishop Nicolson of Carlisle, on his way to London 
in the cold January of 1709, commented with relief on his passage of 
Stainmore "as the snow had been deeply tracked". It is not without 
significance that the Romans sited diis road in such a manner that to 
this day the road remains fairly free from snowdrifts. Easily the worst 
blockage by drifts occurs where the modern road-maker's gendy 
graded diversion leaves the steeper but better-exposed Roman track, 
just on the Westmorland side of the summit. 

The long tradidons of the winter traveller through the eastern 
Pennines are shown again west of Pcnistone on the Derbyshire border. 
Here a rough track diverges from the main Manchester road and 
rejoins it after an exposed hilltop course half a mile beyond ; and the 
Ordnance Survey marks it as 'the snow road'. It was probably used 
as such in the eighteenth century and earlier; diat this route by Lady 
Cross began to be used by wheeled traffic is evident from the account 
of die energetic Lady Anne Pembroke who in March 1656 crossed it 
with her coach "where never coach before". Scotsmen can find 
much of interest in the annals of the adventures of the Edinburgh 
mail from the south; the great roads over the Lammcrmuirs have 
their stories of classic snowfalls (December 1836 notably) and 
twenty miles from Edinburgh the cutting at the top of Soutra Hill 
is a place where again the modern lorry-driver finds much the same 
troubles as his forebears. Snow-screens have helped recently. 

Gale, fog, torrential rain, hail and thunderstorms add their quota 
of incident to the annals of many a parish ; and even tornadoes of some 
violence are not unknown inland. Statistics of the possibilities are 
again better sought elsewhere, but a brief note may be added here as 
from time to time these more violent manifestations of weather also 
play their part in moulding the scene as well as our impression of it. 

All around our coasts winds of gale force (exceeding Beaufort force 8 
i.e. maintaining for an hour or more an average speed upwards of 
38 m.p.h.) blow with some frequency, although as most gales blow from 
between S. and W. the frequency of such excessive winds is greatest 
on our N.W. coasts in Scotland and Ireland and least in N.E. England, 
where the wind speed from any westerly point has been checked over 
the land. Inland however winds attaining gale force for an hour or 
more are in many places rare; although on 29 November 1938 Car- 
dington in Bedfordshire recorded a mean speed of 59 m.p.h. over an 



hour, with a maximum gust of 86 m.p.h. The gale of 16 March 1947 
which gave a gust of 98 m.p.h. at Mildenhall seems to have been even 
more noteworthy as far as damage was concerned; for it is the gusts 
which damage structures and bring down trees. 

The effect of friction is to diminish the average speed of the wind; 
but the fluctuations between gusts and lulls are much more marked. 
If the wind averages force 6 (25 m.p.h.) inland it is quite probable that 
many gusts will exceed gale force, and these are just as liable to cause 
structural damage as if the location were on the coast. An example is 
shown by the ancmogram for South Kensington (Fig. 49, p. 153). 

Suffice it to say that, by and large, the windincss of the British 
climate increases north-westward. It is not easy to give precise 
statistics for frequency of days with gales for any given location. The 
difficulties of comparison even on the coasts arc considerable; over 
such periods of years as are available for the three Hebridcan stations 
Stornoway gives 24 days yearly with gale force recorded at some time 
during the day, Castlebay 11 and Tiree 33. The reasons for these 
differences are apparent when local exposure is taken into account; 
Castlebay provides a relatively well-sheltered harbour, whereas Tiree 
is an island of very low relief. 

Up to 1954 the highest gusts on record ranged from over 70 to 
125 m.p.h.; highest hourly wind speeds from values of the order 
of 35 m.p.h. (force 7) to nearly 80 m.p.h. The latter speed is occa- 
sionally reached at stations such as Scilly, Pendennis Head by 
Falmouth, St. Ann's Head in Pembrokeshire, Southport and Fleet- 
wood, the north-west coasts of Ireland and Scotland, and also Bell 
Rock off the Firth of Tay, Orkney and Shetland. But many windy 
coasts lack an anemograph; inland, too, the variations in average wind 
speed shown by open airfields in comparison with farmed and wooded 
areas arc very evident. Little information is as yet available for our 
uplands. The exceptional winter gales of 1952-3 are noted on pp. 271-2. 
Summer thunderstorms occur more frequently in the Fast Midlands 
than elsewhere, and in the middle Trent valley thunder is heard on an 
average of about 20 days yearly. Torrential rain such as that which 
flooded Louth in 1920, and more rarely, catastrophic hail are often 
associated with thundery conditions. Very extensive damage to crops, 
greenhouses and the like by hail was reported from Kent in 1916 and 
from West Suffolk in 1946; other instances have already been men- 
tioned. Summer thunderstorms arc also rather frequent in some river 


valleys west of the Pennine watershed, where Stonyhurst averages 
about 18 days yearly with 'thunder heard'. (Cf. Marshall, p. 272.) 

Small tornadoes have been reported from time to time from many 
localities; they also arc a concomitant of violent thunderstorms. The 
best described disturbance of this type preceded the passage of a cold 
front across South Wales in October 191 3. In June 1937 a small 
tornado crossed the southern outskirts of Birmingham, blowing out 
doors and windows; on December 8, 1954, another crossed West Lon- 
don and wrecked Gunncrsbury station. Occasionally small dis- 
turbances of great localised violence, probably of tornado character, 
have also been reported for example in Nottinghamshire, Staffordshire, 
Derbyshire and South Lancashire as well as several East Midland 
counties and East Anglia. A very interesting early account of a tornado 
which damaged Widccombc Church on Dartmoor in 1638 has recently 
been given by L. C. W. Bonacina (1946). Exceptionally unstable con- 
ditions, with cold air above over running warm moist surface air, 
appear to be necessary; such conditions are probably more likely to 
develop in spring and summer, but are not of necessity confined to that 
season. Indeed, one of the best defined tornadoes was that at Notting- 
ham on 1 November 1785 [Gentleman's Magazine, 1786). 

The incidence of violent thundery rains and damaging hail storms, 
and even perhaps the chance of an occasional small tornado, are prob- 
ably associated with the regions of most frequent thunder. Small 
scale maps indicating frequency over a considerable period have been 
given by W. A. L. Marshall (1934). 

With further study of the observational material, among which one 
may mention the collected data of the Thunderstorm Census Organ- 
isation under the directorship of Mr. S. Morris Bower of Huddersfield, 
it scorns probable that in time more detailed maps can be drawn. 
There is a good deal of evidence for the view that the frequency of 
overhead thunderstorms (and presumably their consequences in the 
form of damaging hail) shows appreciable local variation; but precise 
observations arc not easily contrived. Mr. A. B. Tinn of Nottingham 
has analysed the incidence of thundery rains in the neighbourhood of 
that city and shows that they tend to be more intense in certain areas, 
notably just west of the city where the valleys of the tributary rivers 
Ercwash and Leen join that of the Trent (Tinn, 1939). Studies on 
similar lines with regard to the validity of local proverbs such as 
"if it thunders at all, it will thunder at Thirsk" — at the foot of the 


Hambleton Hills on the east side of the Vale of York — might well be 
useful. Some will be tempted to suspect from Mr. Morris Bower's 
maps that the broad belt of oolitic limestone running across England 
and including the Cotswolds, may be considerably less liable to hear 
thunder than the valleys on either hand ; but it would be wise to await 
further observations. It is interesting to correlate with this possibility 
an observation by Mrs. Ann Douglas, the well-known glider pilot, 
that at times sailplane pilots have found difficulty in this area from an 
unexpected lack of vigorous thcrmals. The matter, like others, needs 
further investigation. 

The incidence of long-continued rains, and of rainfalls of excep- 
tional intensity, have been the subject of many studies in British 
Rainfall. It is noteworthy that two of the greatest observed amounts of 
rainfall in a day have both occurred in the same region towards the 
foot of the Mcndip Hills in the Somersetshire levels; namely, 9-56 
inches at Bruton (28 June 1917) and 9-4 inches at Cannington 
(18 August 1924). None of our mountain ranges has quite attained this 
amount in a day, although a number of 'orographic' falls exceeding eight 
inches in the day have been recorded. (Lynmouth 1952, see p. 272). 
Thunderstorm rains on the Pennines (compare Chapter 3, p. 52) 
as well as on some of our other hill ranges which arc less visited, have 
undoubtedly given very heavy localised falls in the past. In addition 
to the Stainmore cloud-burst in 1930 previously mentioned, records 
of a flood near Todmorden in July 1870 undoubtedly point to a local 
fall of upwards of nine inches of rain. More recently, severe damage 
was caused at Holmfirth by a similar storm in 1944, causing a sudden 
spate in the streams of the West Riding of Yorkshire. Daily rainfalls 
in excess of four inches have at some time been recorded from most of 
the English, Welsh, and Scottish counties; but such torrential down- 
pours naturally play more part in moulding the scenery in the uplands 
where the run-off is rapid. Professor Austin Miller has commented 
on the results of such a downpour in Wales. (Lynmouth, see p. 272.) 
Spells of excessive rain of several days' duration are more usually 
associated with slow-moving depressions in which the fronts are nearly 
stationary, especially in the summer months when the water-vapour 
content of the air is greater. In 1930 from 20-23 July nearly twelve 
inches of rain fell at Castleton in the North York Moors, and the 
consequent flooding in the Whitby Esk was very great. Other instances 
include the "Moray floods" of 1829 and the extremely heavy and 


persistent rain of 11-12 August 1948 which did so much damage to 
bridges, roads and railway in S.E. Scotland; these have already been 


50 100 

Fig. 68 
Annual rainfall over the British Isles 

mentioned (p. 123). From every part of Britain however there are 
accounts of historic floods; some of the worst have arisen not as a 
result of long-continued summer rains, but from the combination of 



heavy warm rain Calling on a deep snow-cover. The great March 
floods in ihe Fenland in 1947, like those of the Severn were largely 
due to this cause, as an exceptional depth of snow had accumulated 
on all the surrounding uplands and the underlying ground was already 
saturated. Very extensive flooding in both areas befell in February 
1 795 as a result of a sudden thaw with rain. Farther north, apart from 
such Highland rivers as the Spcy, the great and rather sudden floods 
of the lees and Tync, draining a large upland on which both snow 
and rainfall arc liable to be heavy, arc often mentioned; here the 
historic calamity which washed away almost every bridge on both 
rivers befell in November 1771. 



4b 'so do ro 'to 'do iaco '10 '20 So Vo 'so c*> 7b fcb 'so 1900 To So so Ho 5o '60 

Fig. 69 

General rainfall, England and Wales ; percentage of the average for 1881-1915, 
for the years 1727-1960 (from data by J. Glasspoole, brought to date) 

Results of studies of intensity, amount and duration of rainfall and 
of some historic floods have been summarised by Dr. Brooks and 
Dr. Glasspoole of the Meteorological Office and some additional 
figures are given by E. L. Hawkc (194a). Statistical investigations for 
numerous individual stations have been made; in recent years, for 
example, Liverpool (Reynolds, 1953), covering 1 867-1 951 and an 
exhaustive study of Wrexham and district (Ashmorc, 1944) covering 
1 880- 1 942. Ashmore noted that it is usual to find that the number of 
years in excess of the average is appreciably less than the number below 
the average, a fact which will give many food for thought. Over 63 
years at Wrexham the range of variation has lain from 65% to 144% 
of the overall annual average (64 to 145, Isle of Man, Reynolds, 1954). 

1815 '25 "35 45 55 '65 '75 '85 '95 1905 15 '?5 '35 . 45 Man 













1615 '25 '35 45 '55 "65 '75 '85 '95 1905 '15 '25 '35 45 Mean 
Fir.. 70 
Monthly rainfall in inches at Radcliffe Observatory, Oxford, from 1815 
to 1947 (drawn by D. S. Brock) 


The characteristics of the average distribution of the monthly 
rainfall over the British Isles, with its tendency to a maximum in the 
late autumn and early winter months at all more westerly stations and 



i i i i i i i i i i i 





m^'- : •& 

i i i i r i i > I I I 


_ • 

i t i i i i i t t i l 




Fio. 71 
Variability of British monthly rainfall shown for different places and period*. 

Unit: inches. 
The black spot in each column represents the monthly average for the period 
shown: The columns represent the interquartile range (half of all the obser- 
vations above the mean, and half of all those below the mean fall within the 
range shown (after P. R. Crowe, Q.. J. Roy. Met. S., 1940). Note the variation! 
in the Edinburgh pattern for each half-century 

a minimum in the spring are so well known that they will not be 
recapitulated; suffice it to remind readers that throughout a great part 
of the Midlands the rainfall of July and August commonly rivals in 
amount that of the autumn months. Diagrams showing the average 


monthly fall at representative stations can be found in the standard 
work by E. G. Bilham to which reference should be made by those 
who desire an extended discussion of the statistics and of the range 
of variation at individual stations in different districts; and the dis- 
persion diagrams added to the Ordnance Survey's recent map showing 
the average annual rainfall over Great Britain are also to be com- 
mended. Variations in the monthly rainfall at Oxford, 181 5-1947, are 
shown in Figure 70, p. 267, and are characteristic of a large area, over 
which (1916-50) November is the wettest month. 

The wettest districts have already been mentioned, together with 
the frequency of rainfall ; it may be added here that it is considered 
that the village of Great Wakering, near Shoeburyncss in Essex is 
probably in the driest area, with an average annual rainfall (1881-1915) 
of 18-4 in., falling on about 150 days (for 1916-1950, about 19-2 in.). 

The variability of the duration of sunshine from year to year is 
considerable, and it is not directly associated with the rainfall. The 
best source of information is the Monthly Weather Report of the Meteor- 
ological Office. A recent table of the monthly and annual totals of 
bright sunshine at Bognor Regis for 1924-43 was published by D. S. 
Hancock (1944); as it is representative of the sunniest district in 
Britain the averages and extremes are quoted here together with those 
of Durham for comparison. Both sets of figures give a useful impression 
of the range of variation. In 1959 Bognor recorded 2,095 hours. 

Bognor Regis. 1924-1943: Bright sunshine, hours. (D. S. 







66 83 147 166 215 241 218 209 167 117 71 61 

103 127 198 227 279 307 309 271 250 141 104 79 

4* 2 9 9 6 I2 5 '34 20 3 "47 '7 2 Io8 75 47 43 


2066 (1933) 

>5'9 ('932) 

Durham. 1886-1947 (E. F. Baxter, Durham University) 




48 66 107 136 162 176 159 146 124 91 57 42 
79 116 191 238 248 297 253 235 204 151 120 79 
13 32 48 71 87 92 74 57 65 42 29 IO 


1606 (1901) 
982 (1912) 

The Meteorological Office tabulates for a limited number of 
stations the frequency of temperature maxima and minima between 
given limits. The seasonal distribution over the British Isles of the 



number of days with a minimum temperature of'32 or below has been 
summarised by Miss L. F. Lewis (1943). This number averages from 
50 to upwards of 100 at representative inland stations (Cambridge 70, 
Mayfield in Derbyshire 83, Eskdalemuir no, Fort William 63, 
Armagh 49 arc typical) diminishing to much lower figures on exposed 
coasts, e.g. 13 at Falmouth. From what has already been said, 
however, in Chapter 10 it will be evident that very large differences 
exist even between adjacent stations, so that figures such as those 
quoted above can only be taken as a general guide. Particular local- 
ities must be considered with careful attention to the characteristics 
of the site. 

Reference may be made to the aurora borealis — the Northern 
Lights or die 'Merry Dancers' of Scotland. In the north of Britain the 
pale cold yellowish-green glow low down on the northern horizon, 
from which occasional faint streamers dart, momentarily, like in- 
describably distant searchlight beams, upward into the sky above, is 
quite a common sight on clear winter nights; the increase in frequency 
is very rapid from southern England to Shedaud. In the south a keen 
observer far from the lights of towns may detect a faint auroral glow 
on perhaps five evenings of the year. Rarely in the south does it attain 
sufficient intensity to attract the attention even of the countrymen. 
But in the latitude of Aberdeen it may be observed on upwards of 
thirty nights annually and many more in Shetiand; and correspond- 
ingly a greater number arc sufficiently vivid to be widely observed. 
Sometimes other colours are seen, and no one who witnessed the 
magnificent "great aurora" in January 1938 will ever forget the 
wonderful deep red glow shot through by the normal greenish-yellow 

Aurora is due to the impingement of a discharge from the sun, 
considered by many to be formed of a stream of electrically-charged 

Plate XXIa: The humid Irish summer (August). Moist maritime polar air stream 

wilh extensive strato-cumulus and cumulus over County Dublin, breaking a little 

as it debouches over the sea to eastward. Occasional showers inland; dark cloud 

base shadows indicate that the cumulus is rather deep. 

i: A similar type of day in S.W. Ireland (July). Showers in the surface 
air stream, cumulus building up considerably over mountains. Lenticular cloud 
above suggests smooth flowing upper air around the margin of an anticyclone 

centred towards England. 


particles, on the rarificd upper atmosphere, li tends to be most frequent 
about the month of February but varies considerably in frequency 
from year to year and is closely associated with those variations ol 
solar activity revealed by sun spots. The coldly fascinating beauty ol 
a Highland winter night when the aurora flickers above the sileni 
mountains offers to many ol us a variation ol our normal experience 
only to be contemplated with awe. 

Mr. James Paton of the University ol Edinburgh who took 
the photograph on PI. VIII, p. 123, has organised the Scottish aurc :al 
observations; and it is a pleasure to acknowledge the opportunity ol 
showing yet another aspect ol the British scene. 


Chapter 13 

Ashmoke, S. E. (1944). The rainfall of the Wrexham district. Q,. J. Roy. 

Met. S. 70: 241-73. 
Baxter, E. F. (1948). Observations at Durham University Observatory. 

University Offices, Durham. 
Bilham, E. G. (1938). The Climate of the British Isles. London, Macmillan. 
Bonaclna, L. C. VV. (1946). The Widecombe Calamity of 1638. Weather, 

1: 123. 
Bower, S. Morris. Thunderstorm Census Organisation : Annual Reports. 

Published from the Organisation's Headquarters; Oakcs, Huddersfiekl. 
Brooks, C. E. P. and Glasspoole, J. (1928). British Floods and Droughts. 

London, Benn. 
Defoe, D. (1927, cd. Cole). Tour through Great Britain, vol. II, 596-98. 

London, Peicr Davics. 
Douglas, A. C. (1947). Gliding and Advanced Soaring. London, John 

Douglas, C. K. M. (1952). Synoptic aspects of I he storm over N. Scotland 

on Jan. 15, 1952. Meteor. Mag., 81: 104-106. 
Glasspoole, J. (1931). Heavy Falls of Rain in Short Periods. Q,.J. Roy. 

Mel. S. 57: 57-69. 

and Hancock, D. S. (1936). The Distribution over the British Isles 

of the Average Duration of Bright Sunshine. Q.. J. Roy, Met. S. 62: 

Gold, E. (1936). Wind in Britain. Q_. J. Roy. Met. S. 62: 167-206. 
Hancock, D. S. (1944). Sunshine at Bognor Regis. Q.. J. Roy. Met. S. 70: 

Hawke, E. L. (1942). Notable Falls of Rain. Q,. J. Roy. Met. S. 68: 


C.B.S. T 


Hudleston, F. (1930). The Cloudbursts on Stainmorc. British Rainfall, 

ig^o: 287-92. 
Lewis, Lilian P. (1939). The Seasonal and Geographical Distribution 

of Absolute Drought in England. Q.. J. Roy. Met. S. 65: 367-83. 

(1943). Seasonal Distribution over the British Isles of the number of 

days with screen minimum 32 or below. Q.- 3- R°y- Met. S. 69: 

Marshall, W. A. L. (1934). Mean frequency of thunder over British 

Isles. Q.. J. Roy. Met. S. 60: 413-24. Gives a useful map. 
Miller, A. A. (1951). Cause and effect in a Welsh cloudburst. Weather, 

6: 172-79. 
Paton, James (1946). Aurora Borealis. Weather, r: 6-1 1. 
Reynolds, G. (1953). Rainfall at Bidston. Q.J. Roy. Met. S. yg: 137-49. 

(1954). Rainfall in the Isle of Man. Q.J. Roy. Mel. S. 80: 78-88. 
Tinn, A. B. (1940). Local Distribution of Thundery Rains round 

Nottingham. Q. J. Roy. Met. S. 66: 47-65. 
Walker, Sir Gilbert (1930). On the mechanism of Tornadoes. Q.J. Roy. 

Mel. S. 56: 59-66. 

Note on Extremes of Wind and Rainfall in 195s, 1953, 1955 and i960 
Reference to the Orkney gale of January 1952 will be found in the paper by 
C. K. M. Douglas above. This was surpassed by the tremendous northerly gale in 
N.E. Scotland on January 31, 1953. On this occasion a gust of 125 m.p.h. was 
recorded at the wind-generator station on Costa Head, Orkney; and the average 
speed of the wind reached 80 m.p.h. at Lerwick in Shetland. Widespread damage 
to the Eastern Scottish forests resulted; and the consequent "surge" in the North 
Sea with its catastrophic coastal flooding is discussed in many journals (notably 
Weather; Meteorological Magazine; Geographical Journal, for 1953. 

The "Lyntnouth disaster" of August 15, 1952, resulted from a very exceptional 
localised rainfall, of thundery type but lasting for many hours, over Exmoor, wliich 
gave rise to torrential flooding in the steeply-descending rivers. At one gauge on 
Exmoor nine inches fell in this storm; there is, however, reason to believe that 
locally the fall may have approached eleven inches within the 24 hours. This storm 
was discussed extensively (Meteor. Mag.; Weather, 1952). 

In another exceptional outbreak of thunderstorm rains near Weymouth, eleven 
inches of rain fell at Martinstown on July 18, 1955. 

Since this book first appeared, several further papers have been published; 
notably II. II. Lamb on "Our Changing Climate" (Weather, 14, Oct. 1959). The 
present writer has now taken monthly means of temperature back to 1680 (G. 
Manlcy, Metcrological Magazine, 90, Nov. 196 1). The winter of 1684, based on the 
mean temperature of the three months December-February, surpassed that of 1 740 
and thus ranks as the coldest in the past 300 years. The winter of 1G95 was notable 
for frequent and persistent snowfall, and was followed by a cold spring and wet 
summer. Indeed, cool springs and cool disturbed summers were characteristic of 
the decade 1692- 1701 and gave rise to much distress. 



/ am but mad north-north-west. When the wind is 
south I can tell a hawk from a handsaw. 

Shakespeare: Hamlet 

Almost everywhere man and his works form an integral element of 
t our British scene. The observant traveller may comprise within 
his vision such a view as was apostrophised by Wordsworth from the 
summit of Black Combe; from which "the amplest range of un- 
obstructed prospect may be seen that British ground commands." 
From the hills above Ampleforth he can reflect upon the pattern and 
the crops of the great Yorkshire fields ; he can almost espy Vanburgh's 
Castle Howard in addition to the villages of Anglians, Danes, and 
Norsemen; he can contemplate the glory of York Minister and the 
sense of values of that great Yorkshircman Fairfax who saved its 
glass; he can consider whether the adjacent splendid curve of York 
railway station, so finely bedecked with advertisements of diverse size 
and colour, is a stimulus or a deterrent to the Northern mind. Yet on 
another occasion the same traveller's vision will be bounded by seven 
sombrely-dressed individuals with colds in a gaslit railway compart- 
ment dating from 1886, while his sensations are delicately enhanced 
by a recognisable stream of maritime-polar air penetrating the ill- 
fitting window. 

Beneath all these manifestations of British culture lies a climate 
which by virtue of its gentle extremes does not enforce the rapid con- 
demnation of outmoded institutions and equipment. It permits the 
survival, and even the cultivation, of many exotics introduced from 
the neighbouring continent; a statement which may be applied to 
human institutions and ideas as well as plants. But frequently if the 
exotics are not given protection a slow process of adaptation begins. 
Vanbrugh's French-inspired palace became a school. Marxist views 


spread with difficulty away from the shelter of city walls. The Roman 
tide surged but feebly into northern England; when the protection of 
the Wall was removed, the northern peoples quickly swarmed into the 
land of the ash-tree. Mcgalithic man and Mediterranean shrubberies 
show a not dissimilar distribution. 

The influence of climate on man, direct or indirect, defies measure- 
ment. Yet in a country in which man and his works are so prominent 
it is at least desirable to essay an opinion with regard to die extent to 
which climate should be taken into account, in the evolution of this 
integral element of the scenery. Opinions with regard to the ultimate 
effect of climate largely beg the question in a country whose develop- 
ment has been so complex as our own; historians, geographers, biolo- 
gists and anthropologists of every kind will join issue with many of 
the inferences and analogies with which a chapter such as this must 

The physical appearance of the inhabitants of Britain is very varied, 
and if breeding could be applied to them it is probable that widiin this 
island even more varieties would be produced than there arc of sheep. 

As far as we can see climate places no impediment in the way of 
survival of many slightly differing European stocks, although it has 
long ago been suggested that there is a tendency for the fairer types to 
thrive better in the country, so that as a people we are slowly tending 
to become darker with increased urbanisation. Hair and eye colouring 
are often intermediate and only one physical attribute is sufficiendy 
widespread to be, according to some writers, attributable to climate, 
namely, the relatively fair skin with more or less degree of freckling, 
through which the blood-vessels of die face show (ceteris paribus) and 
form what we call a complexion. Such characteristics appertain to 
regions of mild, cloudy and rainy climate in which no marked necessity 
arises for protection of the blood vessels and nerves against cold or dry 
cutting winds; neither is there need for a greater amount of skin pig- 
ment as a protection against the shorter wave-length radiation from 
sun and sky. They are widespread throughout maritime north-west 
Europe, from Bergen to Brest, and may well have given backing to diat 
Venetian Ambassador who reported that England is a country where 
the women have the finest complexions in the world, and the wind is 
always blowing. Yet it becomes questionable whether urbanisation 
with its concomitant, the dominance of fashions derived from cities 
elsewhere, will not lead in this and other respects to a monotonous 



uniformity of style through die deliberate eschewing of that regional 
differentiadon which hitherto has lent so much interest to all the 
means of expression adopted by mankind. 

Some might be tempted to see the slow workings of a more 
favourable climate in the greater refinement of feature characteristic 
of those fair immigrant stocks which have found their home in Southern 
England, by comparison widi the heavier bone so often to be seen in 
the north. But no invariable rule can be made; a walk along Princes 
Street or an inspection of the Raebura portraits in Edinburgh will 
quickly reveal that delicately-chiselled lineaments are no monopoly 
of the men of Kent, or even Dorset. In the fact that our climate 
permits many types to survive with the aid of a little exertion we may 
sec an advantage; cross-breeding of the varied stocks is undoubtedly 
a factor in the production of original genius, as Havelock Ellis and 
others have pointed out. 

Mental traits and accomplishments however show considerable 
regional differentiation. Man's brain is the principal tool by virtue 
of which he survives; and as a tool it may be sharpened or blunted by 
environmental factors. Keenness of perception over a whole people 
can scarcely be measured but it may be observed in the frequency with 
which poetic imagery, artistic accomplishment, and scientific advances 
develop. That the imagery of English and Scottish poets owes an 
immense debt to the country in which they have been bred has been 
evident throughout the fifteen centuries of the use of the English 
language. Such imagery becomes less noticeable as we might expect 
in poets of the town such as Jonson and Pope, but those are relatively 
few. Elsewhere, passage after passage may be found in which the keen 
perception of, and delight in, the sights and sounds of the country 
appears. Throughout our literature associations between the mood of 
the landscape and season, that of the weather, and that of the writer 
can be sought and as quickly found. Can we not allow that these are 
indications of the mental stimulus which testify to one of the greatest 
of the advantages we derive from our variable climate? 

The overwhelming delight in the English spring from Chaucer 
through the Elizabethans, Andrew Marvell's pleasure in his garden, 
the vivid descriptive passages of Dorothy Wordsworth, the upright 
determination of Kingsley's ode to the north-easter are each typical 
of their age. The inspiration born of the mood of the weather has set 
alight the imagination and creative intelligence not only of poets. It 


has captured the regard of the artists who for two centuries at least 
have struggled like the mathematicians with the problems of statics 
and dynamics. There are those who have wresded with the task of 
capturing and stabilising the fugitive delicacy of the moment in the 
manner of Wordsworth's poem "composed upon an evening of extra- 
ordinary splendour and beauty." 

Other men, like Turner, have striven with the rcpresentadon of 
movement; whether in the hurrying clouds, die roaring sea, the bend- 
ing trees, or in the less menacing liveliness of the autumn sowing, the 
pastoral scene, the hunting morning. Can we see in the clearer light 
of the eastern counties not merely the great focus of English 
and Scottish artists, but a place in which, from Cotes and Newton 
to Napier, thoughtful clear-sighted men have logically turned to 

Scientists too have derived at least some part of their stimulus from 
the weather. No better example can be quoted than Dalton, the 
famous proponent of the atomic theory in 1802. Born near Cocker- 
mouth on the margin of the Lake District where water in all its forms 
abounds, the attention he gave for several years in his youth to 
meteorological observations at Kendal bore fruit in the correct 
exposition of the behaviour of water-vapour in the atmosphere. No 
doubt one might add to the inspiration of the changeful clouds that of 
the mountain landscape, where a capacity for taking bold strides from 
footholds already tested is developed to the full. Can we wonder thai 
so many of our most eminent scientists descend from those Northern, 
Western and Scottish stocks bred among the cloudy hills? Especially 
in the realm of physics it becomes quite fascinating to trace the repeated 
associations with environmental factors. It is entirely right to find 
that the need for accuracy of measurement in dealing widi the mani- 
festations of a turbulent Nature is symbolised among the Pennine 
moorland valleys not only by the magnificendy controlled rhythm and 
pitch of die great choral societies. The same necessity has been the 
making of a great line of experimental physicists, inheritors of the same 
sense that lay behind early surveyors and instrument-makers, early 
clockmakers and those great predecessors whose instinct for accurate 
knowledge led to the demand for a Bible men could read : for it has 
already been recalled that WyclifFc and Coverdale were of the same 
stock. Somerset as we saw provides a smaller region in which some- 
what similar attitudes of mind appear. 


The flow of the air, the changeable weather and the rubato 
embroidery upon die fundamental rhythm of die seasons has been one 
of the greatest of the influences moulding the British mind. The charm 
and attractiveness of irregular variations in every sphere of life are the 
greater inasmuch as they are all underlain by the fundamental facts 
arising from our position between ocean and continent. Moreover, 
in a country in which irregularities of pattern of every sort are not only 
experienced but esteemed, tolerance of varieties of opinion has con- 
tinually grown. We have already seen that one of the greatest assets 
of our island, which we can ultimately derive from the climate, lies 
in the diversity of local varieties of plants and animals. Inasmuch as 
the mental attitude in childhood owes much to the results of compara- 
tive observation of surrounding objects and patterns of behaviour we 
can see reasons why local varieties of men tend to differ and why in 
different parts of our island there should arise a predominant "flavour 
of mind". Otherwise it is not easy to see why, through all the mixed 
strains that have populated this island there should appear a per- 
sistent interest in Quakerism and in geological investigation not only 
in Yorkshire but again in Somerset. Likewise nonconformity and the 
biological sciences appear not only in East Anglia but again along the 
Welsh border, associated not infrequently widi that wizardry that 
charms men and makes for political ability — however distorted when 
tinged with the emotions arising from the ergastulum rather than the 
pastures where the wind blows. 

Prudent opportunism too is an immemorial asset; readiness to 
change the plan, to deviate from die policy, to refrain from putting all 
the eggs into one basket may well be attributed to our variable climate. 
Many have long ago pointed the moral — deriving from die addendum 
to many a village notice — "if the weather is wet, the meeting will be 
held in the parish hall instead." But there is also another side to the 
continued beneficent influence of our climate in moulding character- 
istics. Over many years town-dwellers in a climate of litdc extremes 
cease to be affected to any serious degree by the weather. Water 
continues to flow, expensive fruits and vegetables appear from over- 
seas though our own rot, wages arc regularly paid ; the vicissitudes of 
our climate are scarcely more than an inconvenience that upsets the 
football match or the Sunday picnic. Life becomes too easy; and 
perhaps we shall see a changed attitude developing rather rapidly 
when it becomes clear that the food produced in our island is of more 


importance to our welfare than either the seductive surfeit or grudging 
recompense from countries overseas. 

Climate and Livelihood 

In the British Isles, structure as we have seen goes far to mould the 
diversities of our climate. Yet in its turn climate goes far to set limits 
to the use that can be made of the physical features of the country in 
which we live. There are large areas in which the normal agricul- 
tural yield is thoroughly adequate for the maintenance and accumulation 
of energy, a fact well shown not only by the doubling of our population 
in the eighteenth century, but also by the evidence of energy to spare 
for the graces of life whether in the form of meteorological recording, 
tours to the Lake District, walnut furniture or epistolary accomplish- 

But there are also large areas in which practically no material yield 
is obtainable at all. Thirty-five per cent of Great Britain is mountain 
and moorland. Practically all the land above 1,500 feet falls into this 
category and a great deal of it is so poorly drained that only the 
sourest peaty soils are found. Elsewhere on the steeper slopes leaching 
removes enough mineral content from the thin poor soils to render 
them infertile and incapable of supporting more than a poor vegetation 
carrying very little stock. 

In all this it is evident that the amount and frequency of rainfall is 
very significant ; local variations in the degree of slope and the mineral 
constitution of the soil also play their part. In the Lake District 
valleys which drain relatively quickly hay is sometimes cut in places 
where there are a hundred inches of rain; it would be tempting to 
discuss how far this is due to the somewhat exceptional history of 
settlement. In general one begins to find in the Pennines the tell-tale 
abandonment of old reclaimed intakes wherever the annual rainfall 
approaches 55*, and it is highly probable that as many of the enclosures 
were made late in the seventeenth or early in the eighteenth century 
they date from a period when the rainfall may for a decade or two have 
been lower than at present. Farther north this annual average tends 
to decrease. In Galloway unreclaimed moorland predominates above 
the 500-foot contour by reason of the rapid increase of rainfall inland 
as well as distance from markets. Around Loch Laggan and on to- 
wards Newtonmore and Speyside the limits of improvement accord 

climate and man S79 

well with the average rainfall; the like is true in Caithness, allowing 
for the higher latitude. The high altitude reached by settlement and 
cultivation on the Braes of Glenlivet is in agreement not only with the 
greater exposure to light, already mentioned, but also the annual 
average rainfall of 36" at Tomintoul (1,150 feet, in the lee of the 
Cairngorms). Rainfall indeed is more important than temperature in 
regard to the use we can make of our land under present conditions. 

Granted this general princi- 
ple we may also observe that 
there is a considerable area in 
Eastern England on which the 
average annual rainfall is under 
25 inches. The average annual 
evaporation from open water 
surfaces, based on existing data, 
is often quoted as about 16 
inches; but with clear air and 
unsheltered sites at sea level 
this figure is probably rather 
greater. A recently published 
map is reproduced (based on 
calculations given by J. Wads- 
worth, 1948); these data refer 
to tanks in open situations. 
Tank measurements do not 
however give a complete picture ; 
more recently Dr. H. L. Penman 
has given figures for actual 
evaporation from natural land 
surfaces which tend to be 

slightly less than these tank figures, although the principles of dis- 
tribution are broadly similar. It will readily be seen that in a dry year 
the net loss of moisture from the ground will approach closely and 
occasionally even exceed the total rainfall, even in less exposed situ- 
ations, so that irrigation is well worthy of consideration. Percolation is 
also important where the rocks arc permeable. The allowance to be 
made for percolation has been estimated by W. V. Lewis (1943); he 
based his studies on the behaviour of the Breckland meres in S.W. 
Norfolk, some of which dried up completely during the dry years 

Fig. 74 
Evaporation from tanks in open situ- 
ation; average annual amount in 
inches (aAer Wadsworth) 


1933-34. Further, as the evaporation during the summer months 
almost invariably exceeds the rainfall it is broadly true that in much of 
East Anglia the retention of meadows for hay is unprofitable in the 
majority of years. Hence while on the one hand a very large part of 
Britain is too wet for profitable agriculture, there is a considerable 
part in which arable farming for grain must be practised unless the 
farmer is to run the risk of heavy losses through drought. Between, 
there is about one third of die whole surface of Britain in which the 
rainfall lies between 25 and 40 inches. Slight but significant local 
variations in the average seasonal incidence can also be found; for 
example in Cumberland a slighdy greater proportion of the total falls 
in August than farther cast. In such a climate, what is the farmer to 
do? Throughout the Midlands and die lower parts of Wales, N. Eng- 
land and Scodand he must carefully weigh the chances attached to the 
cultivation of winter or spring oats, winter wheat, sometimes spring 
barley, hay, seeds, roots, and potatoes against stock rearing and dairy- 
ing for die towns and cities, near or far. Regional differences in 
agricultural practice are still to a fair extent attributable to climatic 

At best, he and his descendants have become men of shrewd 
judgment and foresight, ready to take a risk on the one hand while 
prudendy insuring themselves with some rival crop; and practically 
every inhabitant of Britain has farming blood within two generations. 
At worst, we must admit dial among our people there are many 
Micawbcrs who continue to hope that "something will turn up" to 
offset losses through carelessness or undue optimism. The British 
climate is rarely so extreme as to lead to catastrophic loss of all crops 
and stock. Suffice it to remember February 1947, when in place of 
the expected reversion to die soft mild intervals of the normal winter 
the snowy cold continued. Those who are prone to believe that the 
British climate is the best in the world should always bear in mind 
that although its extremes may be mildly damaging, they are rarely 
lethal. Western Ireland for long found it possible to develop charm and 
to drift through life in the draughty, smoky, Celtic cottage; a sharp 
contrast with the cleanly and weatherproof Norwegian dwelling, 
demanded by a climate of rather greater extremes. It will for long 
remain disputable which sets of tendencies in our make-up are from 
time to time called out, even in a single life's span, by the varying 
incidence of climatic and economic factors. 



Enough has been said to show that die effect of the British climate 
as an clement in the human environment is more than a tittle para- 
doxical; shrewd opportunism may indeed be plausibly attributable to 
climatic factors, but so may languid carelessness. Becky Sharp, Vernon 
Whitford and Mr. Micawbcr were nearly contemporary. Moreover, 
he who would advocate climatic determinism in such a realm as 
housing cannot afford to overlook the admirable constructive ability 
of languid Malta, or the heedless squalor of die bracing Argentine 
pampa. Many factors beside climate enter into such arguments. 

SEASON at thc eoo-roor mm 


• rc-.-,--i 




Overall Mean 71. Range in aOo years 50 10 92. Wannest decade 81, coldest 58; present 75. 

Fig. 73 

Accumulated temperatures during the growing season, in 'month' degrees, 
1747-1950, for Coo ft. above sea level on the Western 1'cnninos 

In regard to agricultural production it is probable that no countries 
owe more to the efforts of man, through drainage, shelter, and careful 
choice of seed and stock, dian the three realms of Great Britain. On 
the credit side we may safely say that the British climate is one in which 
such improvements can be counted on to give their reward; diis was 
recognised in the sixteenth century, as the quotation at die end of this 
chapter will show. Rarely do calamities attain such magnitude that 
there is widespread ruin. 

The influence of climate on the primary farming activities of man 
in this country is by no means simple. The manifest charms, the 
mental stimulus and the consequent economic advantages appear 
very evident, and have all been mentioned. But some may well ask 
whether a little more extreme behaviour (such as that of the winter of 
1947) would not be a benefit to the country in the long run. At least 


wc should pay more attention to the problems of conserving and trans- 
porting our food and raw materials, as well as those of domestic 
construction and heating. 

With regard to manufacturing industry our climate is probably an 
asset of some value, aldiough it docs not demand technical achieve- 
ments in such realms as air-conditioning. The variations of output 
arising from great heat or cold arc probably less than in many other 
countries; and many smaller workshops need but little winter heating 
or shelter. This as in France is an advantage to the small scale pro- 
ducer. Ship-building and other outdoor industries are rarely impeded 
by weather; railways are relatively easily operated. Domestic costs 
are less than in some countries ; in spite of our wasteful fires, heating 
is relatively cheap, and refrigerators can be dispensed with, although 
they are a convenience in the days of limited access to cool cellars. 
Even Charles I enjoyed ice-cream, and may have been the better 
for it. 

It is not to be forgotten that radiant heat from a bright source of 
some type is probably much more needed than in America, where there 
is a high proportion of winter sunshine. Studies have recently been 
made of the factors affecting our comfort within our homes; from 
which it is evident that a healing system is not used to warm people, 
but to ensure that heat is lost from one's body at a comfortable rate. 
We all generate a certain amount of heat, whether at rest or moving 
about; if we are to remain comfortable heat must be lost at the same 
rate as it is being generated. Loss of heat may arise by evaporation, 
conduction to the surrounding air, or radiation. 

In a very cold climate the objects which surround the body such as 
floors, walls and ceilings are at a lower temperature than the body 
itself, hence there is a considerable loss by radiation to them. If on 
the other hand the air is kept warm, the conduction loss of heat is less. 
Part of the reason for the greater warmth of the air in winter, character- 
istic of American offices and houses — or for that matter modern 
Russian buildings — no doubt arises from this; it helps to balance the 
radiation loss. Much advantage is also to be gained by careful in- 
sulation of walls and ceilings; the doubling of windows is also helpful 
as the air space between is also a poor conductor. 

In the four months November-February many of the great 
American cities enjoy from three to five times as much bright sunshine 
as we do during the same period, and in the lower latitude it is much 


stronger. It is considered by many, therefore, that we have a physio- 
logical need for a high-temperature source of radiation in the form of 
an open fire or its equivalent. 

In the realm of horticulture and gardening wc find that with a very 
little shelter by means of walls, as in many a kitchen garden of the kind 
attached to country houses great and small, much can be done to 
mitigate the incidence of severe frost and to decrease the damaging 
effects of strong wind. Choice of site, combined with the planting of 
trees, has again been the foundation of many a successful nursery. 
Such precautions again lie well within the compass of the individual; 
more so than in many countries where greater precautions are 
necessary. By way of illustration of these advantages we may recall 
how often the English gardener can adequately protect his delicate 
plants in severe weather by throwing a sack over each; in Stockholm 
the countrymen of Linnaeus must devise much more elaborate hoods 
for their carefully nurtured exotics. By contrast we may also remember 
the grateful shade of the many deciduous trees in the Buen Retiro at 
Madrid; but each is surrounded at the base by a stone-lined pit to 
which the water can be led at intervals during the pitiless drought of 
summer. The English gardener often gets his results with his own 
watering-can. May wc not deduce from all this that by comparison 
with many other countries our climatic environment offers especial 
encouragement to small-scale individual effort and enterprise, of which 
the most prominent expression to the foreign visitor is the endless 
variety of suburban gardens relieving the universal monotony of the 
semi-detached villa? 

The astonishing variety of temperate flowers, fruits and vegetables 
that we can with care cultivate has already been noted. From the 
time that the Crusaders brought us cauliflower, men have combined to 
embroider the British landscape with new introductions. Small-scale 
enterprise in yet another respect has thus been encouraged. But our 
climate has its limitations, shown very well by the fact that few 
Mediterranean fruits will ripen, owing to the lack of summer sunshine 
and warmth. The olive baulks at the Cornish Riviera. As for the vine, 
it remains a shadowy ghost along the slopes of the North Downs and 
medieval Cotswolds. Even though our summer climate may have been 
a shade warmer during several decades of the eleventh and thirteenth 
centuries the evidence is doubtful that any wine worthy of the name was 
ever made on a large scale from English grapes. Yet experiments in 


vine growing are again being tried in Surrey and the fact that even 
the pomegranate has been ripened in 1933 and 1949 is noteworthy. 

Climate then has been one of the major factors in the evolution of 
our present landscape, which in England at least owes so much to the 
work of man. The characteristic park-like vistas of the Midlands; the 
varied trees, native and exotic; the hedgerows and irregular enclosures 
all testify to the individualism of successive generations of land 
improvers, following the many centuries when as Trcvclyan has said, 
the sound of the axe in the woodlands was the ground-bass of English 
history. Scotland and Wales are less easily tamed. Yet even in the 
far north-east the stone flags fencing the fields of Caithness and the 
great consumption dykes outside the Aberdeenshire farms not only 
form an element of the scenery; they testify again to a climate in 
which hard individual effort has more often been rewarded rather than 
discouraged. The Aberdeenshire farmer will for long continue to cast 
the loose stones from his ploughlands on the adjacent pile, that he may 
save Ills plough-share; and when the September west wind is drying 
the pale oatficlds of Buchan he may well take pride in his accomplish- 
ment together with the resultant progeny of divines and dominies. 

Climate too has indirectly played its part in the widespread and 
early growth of small-scale industry; the ever-open sea and variable 
but lively air were rather a stimulus than a deterrent to navigation. 
Toynbce's environmental challenge was always present in one form or 
another. Yet we must not lbrgct that it is a corollary of our climatic 
environment that round the corner out of the wind there are numerous 
sheltered places in which the grasses flourish, the animals feed them- 
selves and life for many can appear to be dangerously comfortable 
and secure until the local resources begin to fail. East of Dartmoor, 
South Devonshire has a long tradition of conservatism. 

Climate, Disease and the Death-rate 

Crude statistics of the death-rate do not at first suggest that the 
British climate is extravagantly healthy. But the summer and winter 
temperatures are such that protection against disease is relatively easy. 
The multiplication of fungi and bacteria, the spread of insect-borne and 
water-borne diseases are in large measure affected by air temperature. 
The American visitor is often horrified to sec the meat exposed for 
several hours on the butcher's slab open to the street; he forgets that 




many risks can be taken with a July mean temperature of 58 which 
are quite unpardonable when the mean reaches 75 . 

Moreover the overwhelming majority of our population dwell in 
towns and cities ; if we take account of this it becomes evident that our 
climate is one of the healthiest in the world, largely because the 
sanitary precautions necessary for the continued existence of urban 
communities can be less elaborate, both in summer and winter. 
Severe winter cold in the more primitive communities of Eastern 
Europe leads to overcrowded and uncleanly dwellings among which 
typhus and other louse-borne diseases are endemic. 

Sanitary measures were relatively easy to apply; but before they 
became effectively organised in the middle of last century the death- 
rate of London had become truly appalling by our present standards. 
Not only were eighteenth-century writers fond of commenting on the 
health and stature of many country folk; the development of the 
public schools owes an enormous amount to the instinctive recognition 
that even in our temperate climate children throve better away from 
the town. 

Eighteenth-century physicians were passionately fond of discussing 
the relationship between diseases and the weather; some of these 
habits of mind persisted, in lengthy Victorian publications regarding 
the merits of British watering-places. In early Victorian days we find 
a remarkably widespread interest in meteorological records at our 
south-western resorts, probably deriving Com Dr. Forbes' work on the 
climate of Penzance (1807-1828). Careful digests of observations can 
be found for Helston, Truro, Falmouth, Plymouth, Torquay and 
Sidmouth among other places dating from the age when the many 
delightful little houses were being built from naval prizemoney, and 
from the profits made in tin, copper, East Indiamcn and Brunei's grand 
conceptions of what a railway should be. To this day the tradition 
survives among the many retired naval officers and others who main- 
tain a rain-gauge for the British Rainfall Organisation in those parts. 

With its limited range of temperature and the normal absence 
of persistent steady heat or cold our climate is easier to live in, and so 
imposes less of a strain on our organism. The daily individual pre- 
cautions against disease, which may become an irritation in Egypt 
and a nerve-racking anxiety in Nigeria, are scarcely necessary 
thanks to public sanitary measures of relatively simple and inexpensive 



Yet other ailments beset us besides those devastating epidemics 
against which fences have been erected. There appears to be an 
undoubted association between dampness and some forms of rheu- 
matism, while the drag on our energies imposed by the common cold 
is significant. How much this is due to excessive urbanisation, and how 
much to the initial chill induced by damp clothes, internal or external, 
combined with a penetrating gusty wind in the lower forties, it would 
be hard to say. With all their central heating and consequent dry 
warmth the Americans are similarly troubled. It is by no means 
improbable that the changeability of our weather from day to day 
tends to build up the resistance to such minor infections. At least one 
commentator on tropical hygiene has pointed out that monotony 
tends to lower the resistance, so that occasional departures from the 
normal are often marked by an outbreak of colds. 

Longevity is not quite so great as in the Scandinavian countries, 
and the infant death-rate is by no means the lowest in the world; but 
obviously many factors besides climate must be taken into account. 
It is significant that the general well-being, as measured by vital 
statistics, of the excessively rainy, cloudy and cool Faeroe Islands is 
higher than diat of Western Ireland or the Hebrides; this illustration 
will serve as an adequate reminder of the dangers of climatic deter- 
minism. Recendy the fact that unemployment was greater in Lan- 
cashire than near London in the 1930's was advanced as an argument 
in favour of the climate of the south-east; it would be interesting to 
see whether Macaulay would not have advanced exactly the opposite 
view a hundred years ago. We had better content ourselves by saying 
that at least the British climate as a whole is less inimical than others 
to physical and mental liveliness; and that within it we can perceive 
differences whose explanation may now be attempted. 

Bracing and Relaxing Climates 

So conscious are we of the two opposed effects resulting from our 
maritime climate that we are prone to refer to many places as bracing or 

Plate XXIIln: Breezy August afternoon on the Norfolk Broads; westerly wind 
with fair weather cumulus and cumulo-stralus. 

b: Harvest in East Sulherlandshire. Frontal cirrus and cirro-stratus 
above heralding the approach of more rain; mid-September. 



Plate XXIVa. 


C. L. Cumin 

G. Dough r Bolton 


relaxing in a manner which may well puzzle an American visitor who 
compares the meteorological statistics. Falmouth and Blackpool 
have almost the same July mean temperatures and a very similar 
rainfall. Harrogate and Abernethy record very similar temperatures 
in summer, yet again the Scottish townlet at the head of the Tay 
estuary is regarded as relaxing while Harrogate has an almost universal 
appeal as one of the most bracing inland resorts to be found. Buxton 
lies in an upland valley with a considerably higher relative humidity 
in the summer months than Cambridge or Oxford. But no one 
who knows the sleepy languor of an August afternoon with a 
light south-west breeze at Cambridge will accept the view that 
owing to the greater dampness of Buxton that resort is more relax- 
ing; and any comparison of opinion regarding the local climatic 
merits of British resorts reveals no simple explanation. 

The bracing or relaxing qualities of the air are not yet measurable, 
and opinion regarding them varies immensely among people of different 
age and temperament. Yei after all psychological factors have been 
discounted, there is on many occasions a stimulus to be perceived in 
the atmosphere of some localities which elsewhere is more or less 
lacking. The like indeed may be said of times as well as places; 
"Afflavit Deus, el dissipati sunt" on the Armada medal epitomised an 
age, as well as the joyous turbulence of the fresh west wind of the 
English summer. 

Bracing qualities arc commonly attributed to places where the air 
comes free from pollution over a wide stretch of open sea, moorland 
or mountain; and it is possible that many town-dwellers sense a real 
if intangible benefit from a purer air. In 1687 we find Celia Fiennes 
complaining of the climate of Bath. In 1750 it became a custom to 
take a Sunday walk on Westminster Bridge to escape from the smell 
of the town. Further, there is little doubt that small daily changes in 
the local derivation of the air arc associated with changes in quality, 
notably as regards temperature and humidity and possibly in other 
respects of which we are as yet ignorant. Such changes arc appreciated 

Plate XXIVa: Late summer cloud-cap on Glamaig, Skye; illustrating the results 

of humid air flowing fairly freely over summits. 

b: Midsummer midnight on the north coast of Scotland. Daytime 

cloud has subsided into long streaks of stratus; air nearly calm over the land. 

Distant remnants of cumulus over the sea. 

O.BJ. v 



as a stimulus to the surface of the skin and presumably therefore to the 
nervous system. Variations in wind speed also play their part. It is 
noteworthy, as we saw in previous chapters, that it is particularly 
among mountains or in the neighbourhood of the sea that local 
breezes arc often superimposed on the generally prevailing wind 
system; hence no doubt the so-called bracing qualities of many seaside 
and mountain locations. But this is not the whole story. The incidence 
of the sea-breeze in summer varies considerably with differences in the 
aspect of the coast; and we have already seen (page 158) that its 
scouring effects in the streets of Eastbourne and Bournemouth arc 
likely to be less felt than at Brighton. On the North Sea coast the sea 
tends to be markedly cooler than die land in summer, more so than in 
the Channel ; hence in quiet weather the thermal component of the 
coastal breezes is better developed (compare pp. 156-60 and Fig. 52). 
The North Wales coast resorts arc generally considered to be less 
bracing than those of Lancashire and it will be evident that if the 
gradient wind is on the whole W.S.W. or W., the sea breezes reinforce 
die wind on the Lancashire coast, and incidentally more markedly 
at Blackpool than at Southport; whereas in N. Wales the thermal 
component is at right angles, or even slighdy opposed to the gradient 
wind. Moreover there is some evidence that when the wind descends 
warm and dry from the mountains a sense of relaxation prevails ; the 
warm and gusty south wind, at places such as Forres or Nairn, from 
time to dmc in summer represents the small-scale British modification 
of the Swiss fohn. (Cf. p. 256). Whether Scotsmen are then given to 
displaying even a small measure of Viennese temperament has appar- 
ently not been ascertained. 

Windiness alone does not make a bracing climate. Too much wind 
even if it is merely audible provides for many a degree of nervous 
irritation which may prove tiring; this appears to be especially true 
if the air is extremely dry, as is the fohn of the Swiss valleys. Strong 
wind with a low humidity is however extremely rare in the British Isles. 
More typical are the endless days when the mild and nearly saturated 
south-wester roars unimpeded over the flatter Outer Hebrides and 
Donegal coast; and it must be admitted that many who know these 
windswept fringes of Britain do not regard them as bracing either 
physically or mentally. Speyside is a different matter. Here and 
elsewhere we must also remind ourselves that the mental stimulus 
derived from sojourn in a given locality is not merely a product of the 



climate; it owes far more to the abundance of personal historical, 
literary and scientific associations which the landscape calls to mind. 
Hence an ornithologist such as James Fisher might be tempted to 
ascribe more bracing qualities to South Uist than would a literary 
historian. Discounting such varied opinions as far as we can it still 
remains true that under British climatic conditions the majority regard 
too much damp wind as a deterrent rather than a stimulus to activity. 
Plants, animals and man tend to save energy by adopting the prone 
position. On the long winter nights there is little to do but "sing 
me a song of the lad that is gone", and the beauty of calm days 
in the Hebrides is the more idealised on account of tiieir rarity. 

Bracing and relaxing qualities may well owe a good deal to the 
frequency with which the layer of air within a few feet of the ground 
is changed, for one differing slightly in quality. The rate of change 
with height of the moisture content of the atmosphere is probably 
important and requires further investigation. Under British climatic 
conditions there is generally plenty of moisture in the ground and when 
convection is active, evaporation into the surface layer and the rising 
air currents implies that if there is little wind a man near the ground 
finds himself for many hours in air of much the same quality. The lack 
of stimulus in the humid tropics may thus be explained. But if for 
example on a sunny day in May a dry north-wester is blowing, the 
surface air is turbulent and rising bubbles of humid air from ground 
level alternate frequently and rapidly with descending packets of the 
dry, and physiologically cooler air above. Even in the streets of London 
warm and cool puffs alternate; perhaps indeed they are more frequent 
on account of the different degree of heating on the sunny and shady 
sides of streets of varying width. At least one woman meteorologist 
has verbally expressed her strictly unofficial opinion that London is 
more bracing than the Thames valley above the city. But on days 
when a more humid, tropical air stream prevails, the humidity 
gradient from the surface upward is small. The wind may be quite 
strong and yet the descending bubbles of humid air scarcely allow of 
any more evaporation from the skin than those which are rising. Little 
stimulus is experienced; and as in summer the temperature is often 
high, exertion itself quickly leads to overheating. 

The notably relaxing qualities of a valley-bottom resort such as 
Bath probably derive from the fact that even on a 'bracing' sunny 
day with polar air prevailing the proportion of moist packets of air 


rising from the bottom and slopes of such a sheltered valley is, by 
comparison with an open plain, greater than that of the descending 
packets. Indeed replacement of an ascending packet may be effected 
by a similar humid packet rolling up the valley. It appears that it 
is the combination of moisture with inland warmth and restricted air 
movement that makes for relaxing qualities, and this is borne out by 
the general opinion regarding the Brcckland of West Norfolk, around 
Thctford and Brandon. Here the sandy heaths and pine forests cover 
many square miles; the soil drains very quickly and with a fairly low 
rainfall the area is one of the driest in England. Although it lies a mere 
50-100 feet above sea level many people find the air decidedly more 
bracing than that of Cambridge. 

This hypothesis, that relaxing qualities result from ground moisture 
combined with a high humidity in the surface layers and a lack of air 
movement favouring the interchange from above, agrees well with 
experience in what many deem to be the most overwhelmingly sleepy 
localities in the British Isles. At the heads of the western Scottish lochs, 
and even more those of South-west Ireland, the conditions of a place 
such as Bath or the lower parts of Exeter are exaggerated. The slopes 
arc almost always moist and readily steam in the sunshine; if there is 
any wind to scour away the rising humidity, it must always blow along 
the length of the water so that the surface layers acquire to a con- 
siderable height very much the same damp quality as the air adjacent 
to the slopes. The majority of visitors arrive in the summer or towards 
autumn when gende south-west to westerly winds prevail. There are 
many days when a gende drizzle falls at intervals from the soft grey 
stratus begotten of maritime tropical air, or even maritime polar air 
with a long Atlandc travel; when even that spur to utilitarian activity, 
the Nonconformist conscience, is lulled into a helpless uneasy lassitude. 
Harassed barristers and nerve-racked business men find no better 
cure than fishing under such conditions. 

Another type of day which almost everyone finds particularly 
enervating occurs when in the summer months a 'warm grey sky' 
prevails. Occasionally, with tropical air in particular, much of the 
country lies under a continuous sheet of low stratus cloud above which 
decidedly warmer air is found; in such an event the surface tempera- 
ture is in the neighbourhood of 70°, the humidity is high, and the 
radiation from the warm air above through the cloud-sheet has a 
peculiarly oppressive quality, very different from the wider assortment 



of wave-lengths which we experience on cloudless days of dry air from 
a brilliant sun. It is noteworthy that the enervating qualifies of such 
days increase rapidly with temperature as the diagram on p. 292 will 
show; and that such days tend to be more numerous towards the west, 
notably in Southern Ireland. 

The diagram is taken from the Presidential address by Sir David 
Brunt to the Physical Society in 1947, in which he brought together 
the results of several recent investigations on the reactions of the 
human body to its physical environment. It will be observed that 
with high relative humidities the critical temperature which the body 
finds oppressive lies between 67 and 8o°; above such a temperature 
when we exert ourselves our cooling arrangements are insufficient to 
prevent the body temperature from rising above normal. Even near 
that value conditions are far removed from comfort and a considerable 
sense of oppression will be felt by anyone attempting vigorous exertion. 
It must however, be remembered that the cooling power, that is the 
rate of removal of heat per second by air in movement increases rapidly 
unless the temperature is above 8o°; for example at 70 a wind of 
20 m.p.h. has broadly speaking about double the cooling power of a 
light breeze of 5 m.p.h. and four times the cooling power of calm air. 
But, as Brunt remarks, "with strong winds there will be a tendency 
for the wind to blow through the material of the clothing, and to blow 
through the openings at neck, wrist and legs; it is not possible to 
assess these effects quantitatively." He has also produced a tentative 
classification of climates, shown on die diagram. It will be observed 
that with British temperatures between 30 and 8o° and humidities 
between 20% and saturation, every one of the descriptive adjectives 
can from time to time be applied with the possible exception of 
"becoming irritating" when high temperatures are combined with 
very low humidity. For most British stations it appears very doubtful 
whether the relative humidity ever really falls below 10%. Suffice 
it to recall the many months of raw cold and incessant wind below 40 
which our mountains provide, shown in Appendix, Table III. 

Exposure to the wind at a moderate elevation on a well-drained 
slope is likely to offer the most bracing conditions to the town-dwellers 
inland, a fact confirmed by the siting of many favoured residential 
areas. Hampstcad; Shotover Hill near Oxford; and Alderley Edge near 
Manchester afford good examples. Sheltered valleys debouching on 
to the south and south-west coasts are often remarkably free Com 




a; be a 
f || 5 


u S c C «;-= o> 

1-S-B gs|- 
a 2 ° o § 1 

I g£ J o £* 

« -5 -o .a o ._ B 

J. O S Q. O O 

'*• M a . js — -a .H 

C B 

.9 J . | -9 .9 u 

» 1 5 S .f g -S 

•5.1 § I 

2 J ' C 






1 S 

in 6 2 


8 3 


' 5 J 

Si 5.8 ■ 


S £ o j: « fe 

v B B 99 & 

AJ/O/ftOH 3AUV13U 

■a 1 f | I 



frost, but are considered to be relaxing as we have seen especially when 
the hills on either side are high. In North Wales the Clwyd valley 
round Denbigh has a similar reputation. Open flat coasts adjacent 
to our cooler seas, over which sea and land breezes can play freely, 
are often regarded as bracing. Elsewhere, much depends on soil 
moisture; the clayey Chiltern ridges arc not generally considered to be 
as bracing as the sandy uplands of south-west Surrey. In spite of the 
relative lack of sunshine and the high rainfall many of the Pennine 
industrial towns at higher levels are remarkably bracing, and here it is 
probable that the small daily exchanges between hill and valley play 
some part. Coupled with this we may recognise another factor. The 
arrival of warm humid Atlantic air in summer is commonly associated 
with a peculiarly relaxing effect at many inland towns. At Cambridge, 
an afternoon maximum of 75 with humid air, a dew-point between 
6o° and 65 , patchy cumulus and vigorous convection in the sunshine 
is characteristic. But nearer the approaching Atlantic depression the 
south-westerly wind often blows more strongly on the northern up- 
lands; and even with a dew-point as high as 60 ° a force 5 wind in the 
suburbs of Huddersficld gives appreciably more cooling effect to the 
body than the force 3 breeze of the East Midland afternoon. More- 
over, there is generally more cloud and less sunshine under such 
conditions, so that when the temperature is 75 in the South Midlands 
it is probably not more than 67 ° in the streets of Bradford. Under 
conditions of high humidity every degree makes an appreciable 
difference to human comfort, as the diagram above will again show. 
Hence we find that in the same air-mass a mere six hundred feet of 
elevation combined with rather more cloud and wind make for a 
considerable difference in the comfort of those engaged in active 
exertion, whether work or play. This goes far to explain the per- 
ceptible sense of increased vigour which many visitors readily admit. 
In all this we have said little about the stimulating or relaxing 
qualities of the air in the cooler months. Strong winds even at a moder- 
ately low temperature remove heat from the body more rapidly than 
any other agency normally met with. The cooling power of a 30 m.p.h. 
wind at a temperature of 40 is as great as that of calm air at a tem- 
perature of 15 . At lower temperatures it makes little difference 
whether die air is dry or damp. Over the range from 35 to 50 or so 
dry air cools us more rapidly than moist air; hence the cutting qualities 
of the dry keen north-easter of spring even though its temperature may 



approach 50 . Damp air in the thirties owes its peculiar raw chill to 
the fact that evaporation from the surface of the skin beneath the 
clothes soon leads to saturation of the adjacent air and hence a moist 
layer in the garment next to the skin. The moisture forms a good 
conductor of heat from the body to the cool air outside and unless 
further thicknesses of clothing are worn a sensation of chill quickly 
develops. Wool in a damp climate has the advantage that it is absor- 
bent and that the addition of moisture does not increase the con- 
ductivity so rapidly as would be the case where the fabric docs not 
enclose so much air between the strands, for example cotton and its 
derivatives. Wool moreover is not airtight, a manifest advantage to 
those who exert themselves under humid conditions unless the tempera- 
ture is far lower than we normally experience. Oats and sheep are 
an essential accompaniment of the north-west British climate; por- 
ridge for breakfast and wool next to the skin however unfashionable 
should remain the portion of those who would enjoy the British 

Clothing and diet, gardens and literature, parklands and pleasure 
resorts all form part of the British scene derived from the work of man, 
yet it is evident that the characteristic development of all these 
elements in our landscape has repeatedly been affected by climate. 
Indeed we have essayed the view that these emblems of British culture 
would not have taken the form they have were it not for our climate, 
which continually shows us that at one place or time its effects are 
stimulating and persuasive, while elsewhere on another occasion they 
are relaxing and permissive. We may ask ourselves whether the varying 
predominance of the associated mental traits at different times in our 
history cannot also be recognised; the cheerful and rather casual 
acceptance of Nell Gwynne, "let sleeping dogs lie", and the era of 
appeasement contrasted with the decision implicit in Cromwell's 
Navigation Act, the defeat of the Armada and the emancipation of 
the slaves. 

Twenty years ago Professor Ellsworth Huntington of Yale expressed 
the view that the climate of South-east England might even be more 
advantageous than that of Yale with regard to the development and 
progress of civilisation. We now cast doubt upon his somewhat crude 
basis of assessment, yet the fact remains that a civilisation based on 
human effort is likely to thrive better in a region where both effort 
and forethought are steadily demanded and rewarded. In many 


lands they are discouraged, whether by natural calamity, prolonged 
uniformity of heat or cold, the rivalry of other forms of life, or by the 
more subtle results of soil exhaustion which again is largely a matter 
of climate. Such territories can only maintain a civilisation with the 
aid of large-scale technique, applied by some form of collective effort 
under a management with all that it implies. The battle for main- 
tenance of lively modern communities in the Missouri valley, or 
in that of the Yenesei, is only beginning; a century hence we 
may be able to estimate the permanence of such extensions of the 
civilisations derived from the Western European nursery of small- 
scale individual effort in which all our technical accomplishment has 
been initiated. 

Under the alternation of irritating blandishment and kindly asperity 
which the British climate provides we are at least secured from soil 
erosion; and every feature of our native environment is conducive to 
diversity, deviation and individual differentiation in plants, animals and 
men. Combined with an agreement to differ we find co-operation for 
common ends, in legal matters, in defence and occasionally in develop- 
ment. It is only in the man-made environment of our cities that a degree 
of monotony has been developed which is potentially similar in its 
effects to the Russian plains or the American prairie. It remains to 
be seen whether this group of islands, in which diversities of climate 
and structure have played so large a part in moulding the attitude 
of mind, can possibly be administered by methods and systems be- 
gotten of cogitation in more uniform lands, however greatly their 
logic may appeal to that large proportion of our population which has 
for generations been removed from the sublimely irregular complexities 
and subtle adjustments so characteristic of the country we know. 

Whatever be the result, oxymoron as a figure of speech will con- 
tinue to be appropriate in descriptions of that paradoxical British 
climate which defies definition and in which squalid culture, orderly 
disintegration, and untidy neatness have repeatedly been found by the 
bewildered observer from abroad ; sufficient use of the figure of speech 
has been made already to justify the assertion that the critic of 
British institutions may praise or blame the climate as he wishes. 
No precise quantitative evaluation of its effects capable of satisfy- 
ing a scientific inquiry can yet be made. No precise equivalent 
can be found elsewhere in which we can study the reactions of an 
immigrant community. 


New Zealand is free from the effects of a continental air supply in 
winter and has more powerful sunshine. Vancouver Island at its 
southern end has similar temperatures but a drier and more settled 
summer. Tierra del Fuego is altogether cooler, though generally drier 
apart from the mountainous westward fringe. 

The fundamental advantages of our climate lie firsdy, in the 
encouragement it gives to the development of local differentiation ; the 
consequent variety of objects composing our scenery in a short distance 
cannot fail to act as a mental stimulus. Moreover, such contrasts lie 
well within the reach of die poorest individual; they are not a matter 
of expensive travel. Secondly, our climate imposes a relatively small 
tax on human energy. Since glass became common after the Refor- 
mation, such protection as we require has lain for the most part well 
within the compass of individual enterprise and conscience. Drainage 
of the farmlands and improvement of tillage followed; with die result 
that over the last three centuries there has been a widespread release 
of energy for other purposes. But such advantages were also enjoyed 
in Holland and the other countries of north-west Europe; and it must 
not be forgotten that the potential demerits of the British climate are 
exemplified in those regions of Western Ireland where nascent en- 
thusiasm is too easily damped down, from which the younger generation 
continues to migrate while the older generation recounts the stories of 
a long-past heroic age associated not only with invasion but also with 
diminished rainfall and a keener air. Britain may well be but one 
stage removed from a like fate. Further, who can fail to recognise 
the immense drag upon our productive energies imposed by the 
maintenance of our obese capital city, whose size has so far outgrown 
our resources — limited again as they ultimately are by our climate? 

It must rest with future scientists to devise methods of measurement 
of qualities such as climatic stimulus, productivity of original accom- 
plishment in arts or engineering, loss of energy due to friction from 
overcrowding. Eventually a balance might be struck between the 
expenditure of human energy necessary for continued existence and 
the gain due to natural causes. When that has been done we shall be 
in a position to declare whether die British climate is ultimately an 
asset. At present many who judge by the historical and geographical 
results, of which an indication has been attempted in this and earlier 
chapters will be tempted to the opinion that the astringent mildness of 
our dominant maritime-polar air is one of the few unalterable comforts 



we possess, however much we may personally deplore its harassing 
benevolence. Truly the air that reaches us, in the words of Shake- 
speare's Ariel 

" suffers a sea-change 

Into something rich and strange." 

Yet widi Harrison the Elizabethan topographer we can also agree:— 

"We have if need be sufficient help to cherish our ground withall, and 

to make it more fruitful neither is there anything found in the 

aire of our region, that is not usually seen amongst other nations living 
beyond the seas". 


Chapter 14 

Brunt, D. (1947). Some physical aspects of the heat balance of the 

human body. Presidential Address: Proc. Phys. S. 59: p. 713-726. 
Brooks, C. E. P. (1950). Climate in Everyday Life. London: Benn. 
Ellis, Havclock (1927). A Study of British Genius. New edition, revised 

and enlarged. London, Constable; chapter II, p. 36, 40 el seq 
Fleure, H. J. (1951). The Natural History of Man in Britain. London: 

Collins' New Naturalist. 
Handisyde, C. C, (1947). The Climate of the Home. Weather, a: 82-88: 

gives further references to recent work. 
Hooker, R. H. (1922). The Weather and the Crops in S. England, 

1885-1921. Q. J. Roy. Met. S. 48: 115-38. 
Huntington, Ellsworth (1924). Civilisation and Climate. 224-26, 229. Third 

edition, New Haven: Yale Univ. Press. 

(1945). Mainsprings of Civilisation. 384. New York, John Wiley; 

London, Chapman & Hall. 
Lewis, W. V. (1943). Some aspects of percolation in S.E. England. 

Proc. Geol. Ass. 54: 171-84. 
Manley, G. (1957). Climate fluctuations and full requirements. Scot. 

Geogr. Mag., 73:, No. 1, 19-28. 
Normand, C. W. B. (1920). The effect of high temperatures, humidity 

and wind on the human body. Q_. J. Roy. Mel. S. 46: 1-14. 
Stone, R. G. (1941). Health in Tropical Climates: in Climate and Man. 

Tearb. i/.S. Dep. Agri. (1941): 246-61. 
Wadsworth, J. (1948). Evaporation from tanks in the British Isles. 

Weather, 3: 322-24. 

*»- 75 
Location of places men- 
tioned in the approximate 
order of reference in the 
text, and excluding the more 
familiar ports and cities (tec 
opposite page) 

List of places referred to in the text— with reference numbers as 
given on the map 

I. Durham: Ushaw, 

a. Crossfell: Moorhouse 

3. Burnley 

4. Lowestoft 

5. Lyndon: Stamford 

6. Plymouth 

7. Upminster 

8. Cambridge 
Tolland Bay 

12. Ben Nevis: Fort 
Kendal: Leven-Kent 


18. Blair Atholl 

19. Lunds: Aisgill. Gars- 


Levcn-Kcnt: see 14 
Houghall : see t 

22. St. Ann's Head 

23. Cardington-Bedford 

24. Carlisle: R. liden 

25. Carsiairs 

26. Lindsey 

27. Holdcrness 

35. Blackpool 

36. Buttennere: Keswick. 









37. Harrogate 

73. Queensbury 

38. Wrexham 

Stamford, see 3 

3g. Glcngarry-Glcn- 

74. Huddcrsfield 


75. Snowshill 

40. Athcrstone 

76. Cold Ash by 

41. Shrewsbury 

77. Maiden Castle 

42. Norwich 

78. Fairsnape Fell: 

43. Bournemouth 


44. York 

79. Kinder Scout 

45. Lincoln 

80. Dungcness 

46. Oxford 

81. Stonyhurst, see 78 

47. Whitby 

81. Gordon Castle 

48. Tccsdale: Wcardalc 

49. Braes of Glcnlivct 

82. Callerirk 

83. Great Wakcring 

50. Yarmouth 

84. Bognor 

51. Hartlepool 

85. Black Combe 

52. Margate 

86. Ampleforth 

53. Southport 

87. Carnedd Llewelyn 

54. Bolton 

88. Kelso 


89. West Linton 

water, see 36 


55. Rickmansworth 

sec 16 

Ushaw. see 1 

90. Peebles 

56. Pcrdiswell-Droitwich 

91. Appleby 

57. Malvern 

92. Balmoral 

58. Bromyard 

93. South Farnborough 

59. Breckland: Milden- 

94. Cheadle 

hall, Lynford 

95. Maidstone 

60. Alston 

96. Barnstaple 

Moorhouse, see 3 

97. Wealdstone 

61. Dalwhinnie 

98. Peterborough 

62. Plynlimmon 

Mildenhall, sec 59 

63. Ingleborough 

99. Reading 

64. Dalnaspidal 

100. Tonbridge 

65. Falmouth 

101. Halsicad 

66. Cardiff 

102. Raunds 

67. Greenock 

103. Bawtry 

Fort William, see 12 

104. Achnashellach 

68. Stornoway 

105. Penzance 

69. Stainmorc 

70. Braemar 

106. Ountlle 

107. Marlborough 

71. Prcstwick 

108. i'.iith 

72. Consetl 


03 O 

O 10 


- Ci 

■* tO 





Tf « 

o o 


m o 
•* - 

r» eo 


™ 0*. 

m eo 

CO ei 

eg d 

ei ei 

h. ° 

m - 

■* - 


.0 -a 

in -• 

<n o 

— c 


Ol ■«• 

m — 

% 2 

2 .5 

CO ei 

m to 


eo in 

« c« 

0. S 

to - 

IO - 

w — 

«■ « 

Ol <o 

to 5 

— ■) 

■* in 

« eo 

!• 8 

to - 

m - 

> U 

en - 

m ci 

8 "B 


<8 "2 

3 i* 


oi eo 


s £ 


tO ID 

ffl ■* 


in - 

* - 

.— 1**. 

"• co 

•n ei 

$ 9t 


oi rr 

CO c< 


■*• - 


-0 = 

r» — 

m <o 



3 2 1 

3 » 

re to 
1- u 


•*• o 

> *" 


- O 


<o *n 

TT " 


- *• 

■n « 

to u 
3 3 

■— i 

- Ol 

M f» 

c r 



c to 

< ~ 

. 1 







■§3 '5. 


"c? £ 







— --^ 


<a «, 







A i~ 

*■— • m 

3H J § 


3 •■» x '3 
IS 8f 

Jr c.— 

1 s 




to co "K 



CO o 



to ei' 





r-- _ 




"f _ 

- co 

tO CI 

O Ci 
«1 - 






CO ct 

co t^ 



* - 

? co 

"2 R 


to CO 

01 r~ 
- « 

co ■* 


to ci 


i X 





— : 

tO co 



S'"* - 




r« c< 



o . 

U. CO 






" CO 

en a 




o « 

-to r» ci 
r* ci ci •-< 

- co 


CI * 

CI * 


!9 , 



'f- o 
co eo 






tO CI 

r~ in 
01 CI 


2 c? 

>. CI 


co m 



I rf 

*^ >~ 
§ * 























•* 1^ 

11 ■" 


co to 


to & 

i! m 




J 10 

3 J. 


y O 


S hi 


•a "8 

8 J3 



C 3 

n * 

a a 



O 13 

* « 

k. > 


= ii 

to S 


•5 & 




?■ S3 








O - m 
ei 1 




01 » e< 



O Oi 






a 2. • £. 
r* 01 O". 01 






it". C CI 

01 01 01 



CO to 

ec co 







*o- m m cd 




ci m co 


CD r^ 




CO co to r^ 




CO - CO 


* CO 




- to r~ p 
to b 01 



r~ eo m 


m to 
cVi b 








* O ci 01 




to eo 


9 ? 






co - 01 o-. 

•4- Th CO CO 





- CO 



Ol * CO - 




<r> in •# 


in eo 






in co ci co 









t « Ct CO 




t^ p- 01 


CI 01 



O 0". CO O) 

m f •* ■* 




in m m 


01 b 
•* m 



to to "O- to 




ci - m 


i~ — 



~ — — \o 




N 01 N 


m m 




m in in in 




in in in 





O*. m v r^ 




O h « 


m co 






to CO CO O 

in m m to 




£ 5 to 



cb 61 
m m 



m ci v 



m eo 


01 m 




10 cj 51- 
tn in in to 




to to to 


cb 01 
in in 





t- - CO I-» 



eo co ei 


- to 

- 0, 
2 c 


•8 8 





eo to m f- 
m m in in 



O 00 o-. 
to m m 



in in 



01 m 01 in 




m t O 







01 - 01 CI 







- CI 

m m 



Ol tO CO Ol 




co r- 







in in *** to 
¥ ■♦ * ■* 




r» r~ eo 


to to 

■a- •*• 






« r< Oi 


f-* — 





* 1- •* - 




_ _ CI 





s « 








— ~ X c. 

* n co co 
m to to r» 





*■ 5- *«■ 
■»• 01 - 


O eo 
»»■ * 

ei in 


| — 1 



•» 01 r~ co 
* m n to 




01 - 

eo ■* •* 








o> CO Ol Ci Cl CJ1 

a 31 

u r 

— a — 

cr. c cr. 

f^ r* Q 01 

ei ■»»• v •«• * 

_ii - - to 
01 01 01 01 01 

Ol Oi 
CI ■* ■* TT 

1 1 1 1 

- IO - - 

— ~. - ~ 

01 Ol Ol C". 

"0 2 



C •"' 










h x a ^ 























1 1 

1 f 

60 C C 


ei « a 





TJ ^ 
c ^e 

1 t; 


^n — - 

« § 











nf2 » -1 ~l 

Sh 10 t- 




Ol tl «1 - 

* ib !r> in 

ci r^ « if 
cb ~ b b 



ica is n co 

iO 10 10 OS 

eo co co en 

r*» co o O 
eo co 



* %%■ 

- - r- Cl 10 


us io i*» 



o o - 


CO - - p> ;»• 

<t into fs r* 

io in id tfi us 

I^ CO CO O 10 

•a- io 10 co 
in *o io o 

« i-> m co 





m us 




IO 10 




r- co 



* ■* * ■* 

- O r^ 10 cp 

w « ^- co 

-*•■>*•■* * * 

~ - *r - en 

10 CO CO o 

eo co co *? 
— O IO CO 

<*• to in n 












o ci o m - 

co m m co 

















en t» r-» en 

■«• ** •& rp 

I I I 

- - io en 

co - « - 

en en en en 

o en 
en en 


co o co r-- 
r» c* ^? io 

- eo co r» 


















t** — 



•* i0 




»* ~ 










^ • 

g i 


■^ jj 





■0 *"• 
c - 


w :• 

r-» 10 


O . 



















eo en 







"=> E 



| - 


















as d 








" g 


2 S 

■° s. 



> 5 









os 9 



r^ us 









eb b 

CO - 













— 01 



■0 7 



9 « 








- Ct 

* - 





co r^ 








10 CO 



us 9 











r~ eo 


CO « 
us - 










eo 9 



IB us 




m US 









CO 10 

in - 



% 2 



« en 








? e? 



US 10 


!*" V 








cb in 
•o - 


- CO 

10 - 




ir a. 















w — 


r^ ci 









in 10 
m - 




i. CO 

10 - 










w CO 



« 7 



OS r~ 




— > 






10 cb 
m - 


eb ^ i 



b bs 









r^ <fi 




OS r- 




■? r 







b r^ 

m - 


co co 
us - 



ib en 


i 5 " 




?" 7 







en eo 







in 10 




eb eo 
•* — 




3- * 










?• *? 




us eo 



us CO 






b eo 



in » 



%. * 







r»- r» 



O! CO 



T ■? 









cb - 

CO - 



co b 




CS r** 




ir ; 




r *? 



OS 10 




w. CO 







r- b 

CO - 



eo b. 




eb to 




—— . 



S en 







5 r^ 

i J. 






^ ( 
















■X3 - 

1 C 












C£ u 




' c s 


J2 3 




a £ 








co to * «■» ♦ « ?"' f- ?* » f" *P ?"P •? ^C^i" ? J* J 

* c? « a> m to K co co s c. ■*•*« c. co <n co ■* c- ■*• 


wo o •* o wf^c ci r~ ci Ci **• to to cp p. fl « ^ £"* 


ocnei ij- co f» to n to ti - n n« ci cow ■«■ in m >*• 



o co *■ o as -o - o ei oip f» 7 f » « n 7 c w 7 
ntoii co ci to to n Jiti ci * « w « *•«•*•*« * 



moto-dctrj-ocot-oopcti^r-cccc ;f ? =p 



;»■«« ^-cof^ib encoa « CrsCn ci ci n « m * m on 


ai i> eo n * a> co to co » r~ r» 7 p f> ? K~ S" V" °F 






eoiO r~ ~ O m ci a> ;*• cr p r» so tfjwcp ci 7 p y O 




lomoieo n 01 ci to « ?> tP f T*" *? r 5 '■? S* 7 * 






fc C-. ci co •* ci •«?■ to to inr^r-eptp 7 7 j»-c< 





•*oo mciio - nmco i.-tp p 7 pip p- * «- *• to 





9 ts « i^to coo oco J» p c~ cc - cijf', n 7 l o 

net •• « -nnB---«n--«-i'«« H 




r^ » *■ ■* O -to O c-tJ-ci cvcp r» pi r» co ci «r 


«--««*«»«-- ci ci--ci-eieoe«co 


> .. 


cso ■na.cocom^eoco eoraepcpcp ;»• ep p> uo * CO 





co co ~ co +nni9P«oio*nifi(ipoin 


■— : 

* ci « * co f-. to co co ci ei * in ci ci coci * m en * 









— inO'^mo-'fcomiO-cti^mcior-ciinifiO 



SO < * E Siz 













.1 j I 

a. p 09 S 

Stg ft. " 





g o«S>-="beM0 c 2haS M 

||-5ig1§8a|iliJJ88-g ! S3g§ 1 S 

3o. r ) 



* £ 


o w in 

a £ - - 

30 X :T. Cn 

52- 52. 

c c 

C C 



CO t/3 

M N PI Cn 



CO v 00 - « 

r* r» 10 0* 

c o 


«r> cr> _ 


6 6 

in r-» r* « 


b c b b 


b b o 























S'l 8 


Figures .n heavy lype are pages opposite which illuitrauons are to be found. 

Aberdeen, 43, 114, 130. 145, 
151. 186. 202. 255, 257. 
259, 270, 284 

Aberystwyth, 126, 136 

Admiralty Pilot, 21 

Admiralty Weather Manual. 

Agrotus, 139 

Ahlmann, H. W., 249 

Air Currents. 47, 51; ,,, 

Air Masses,';^, 4°. 7olf, lit, 

119; set Circulation 
Air Streams, 42, 43; Tropi- 
cal, 63; Polar, 63; see 
Aiigill, n 
Alder, 22b, 227 
Aldcrley Edge, 291 
Altered Oidllaiion, 226. 227. 

Mlestrec, 51, fog 
Alps, 214. 216, 239, 241 
..I ten 175, 179,257 
Altitudr, Effects of, 1781I; 
and Frost, t86ff; and Rain- 
fall, igoff, 199 
Amplrlbrth, 273 
Anemometer, Cup, 16; Dines, 

ut Dines Anemograph 
Anglesey. 73 
Angstrom. A., 249 
Angus, 138. 150, 206 
Annual Rainfall Map. 265 
Antarctic, 214 

Anticyclone, 63, 82-91, 163, 
234 ; and Spring Weather, 
"3- "7, 118; and Sum- 
mer, 125; >« Azores Hii<h 
AnticycJonic Gloom. 86 
Appleby, 257 
April, 210 

Arctic Air, 33, 71 ; Maritime, 
78-9, 103, 119; Contin- 
ental, 79, 8t. 88, 119 
Arctic Ice, 119, 124, 125 
Argentine Pampa, 281 
Argyllshire, 145, 237 
Armada Summer, 240 
Armagh, 16, 19;, 
Asugill Plantation, 179 
Asnrnore, S. I... 266. ^71 

Aspect. 193 

Athcrstonc, 140 

Atkinson, J., 8 

Atlantic, 63, 72. 83; ix> wv 
103. 123 

Atlantic Phase. aa3 
Atmosphere, Properties ol, 

August, 123-4, 210 
Aurora Borcalis, 270 
Autumn. 125ft 
Avicmore, 179 
Ayscough, 10 

Azores 24. 59; High, 59. 63. 
6 4> "7. 113- 1 -A 231. 235 

Backhouse, 1., IS 
Bakcr.J. N. L., 242. 240 
Bftlctun, W. G. V„ 249 
Balmoral, 208, 257 
Baltasound, :-/..' 
Baltic, 216, 227. 23b 
Banff, 78, 151 150 100 
Barker, T.. 8. 16 
Barnstaple, 140. 255, 257 
Barograph, 63 
Barometer, 63 
Bath, 287, 289 
Bartholomew's Atlas 01 

Meteorology, 02 
Barton, 209 
Baxter, E. K. 271 
Bay of Biscay, 22 1 
Beaufort Scale. 16, 61 
Beck, 15 
Bede, 7' 

Bedford College, 175 
Bedfordshire, 260 
Belasco, J. E., 92 
Belgium, 29 
Bell Rock, 152 
IVn Nevis, 12, 45, 202. 210. 
-■ r 1 : Observatory 188. 
180, 207 
Ben Wyvis, 210 
Bermuda, 59,83, 919 
Bewcastle Fells, 192 
Bcrwickslure, 257 
Bigclstone, H. J.. 260. 271 
Bilham. E. C, "14. 104 269 

Birch, 234; 98 
Birmingham, Mrj 

Blackpool, fit. 120, 1 4.1 25;, 

Black Wind 01 W. Ireland 

Blacnau Fcstintog. 192 
Blcasdalc. 192 
Blizzard, 104 
Bodiam. Sussex, 88. 1 70 

:nor Regis. 269 
Itolion, 159 
Bonacina, L. C. W., 93. 104 

210, 212, 271 
Bonn, 60 
Borden, XI. 4 

Boreal Period. 228 

Boreas, 2 

Borrowdaie. 16 136, 190. 

Boston, Mass., 69 
Botley, C. M., 54 
Bottom Wind' ol Denwmt- 

water 150 
Bournemouth. 141. 157 
Bowct. S. Morris, 2G3, 264 

Bowfrll, 210 

Bra tag Air. 15H, 286. 287 

Bradlord, 293 
Brad well, Edge. 150 
Bracmar 20 103, 202. 203 

207, 257 
oracnacii, 2 1 1 
llraes of Glenrivei. 193. 194. 

257. «79 
Bra he, I'ycho, 241 
I hast Clough, 195 
''.reckland. 102. 171 290; 

Mere.. 2711 
Brecon Beacon-., 108. 221 
Bredon Hill, tOf! 
Bridlington, 170 
Brighton, 80. 142. 157. ir,8 

Bristol Channel, 62. 221 

ISrihim'.' Structure anil Scenery. 

British Rainlah Organisation. 

t2, 247, 285 
Bromyard, 167 
Bronze Age Summed jm 



Brooks, C. E. P., 23, 54, 90, 

92, 132, 133,215,249,266, 

271, 297 

Brown Cove, 2 to, 211 

Brunt, Sir David, 24, 53, 54. 

55. 9=. '75, '76, =9'. a 49, 

Bucnan, A., 90, 91, 92, 196 
Bude, 169 

Built up areas, and temper- 
ature, 173, 174; and fog, 

1 7; "> 
Bundling, 198 

Burgundy, I 

Burnley, 7 

Uuticnncre, 122, fi9i 

Buxton, 187, 287 

Cairngorm, 145, 179, 188. 
193, 2 to 

Cairntoul, 190 

Caithness, 78, 236, 279, 284 

Calais, 80 

Californian Trees, see Growth 

Cambridge, 12, 18, 44, 62, 
75, too, 101, 102, it8, 
122, 140, 142, 160, 171, 
185, 209, 236. 255- 256. 

257. 270, 293 o 
Campbell-Stokes Recorder, 

'7 . 
Cannington, 264 
Canterbury, 88. 170 
Carncdd Llewelyn. 210 
Cardiff, 194, 268 
Cardigan Bay Dyke*. 238 
Cardiganshire, 178, 255 
Carlisle, 8 

Carr-Laughton, L. U.. 049 
Carruibcn, J., 92 
Gary, John, 10 
Casartelli, 10 
Castlebay, 262 
Casdcton, 264 
Caticrick Bridge, 260 
Cave, C..J. P., 12 
Charles, I, 282 
Charles II, 3, 120 
Cham wood Forest, 140 
Cheadlc, 257 
Chelsea Floods, 237 
Cheltenham, 167 
Cheshire, 61 
Chilterns, 33. 62. [09 

'85, 293 
Circulation, 57ft, 234 
Cleveland, 78, 150, 200, BOS 

Climatic variations, 21 3ft 
Climate Through the Ages, 

Cloud, 34-41; Streets, 52-3; 

Cirrocumulus, 39; Cirro- 
stratus, 37 ; Cumulonim- 
bus, 38, 41-50; passim, 77; 
Slratocumulus, 73; Cloud 
illustrations, 2, 50, 127 

Clwyd Valley, 293 

Clyde, Firth of, 62 

Clyde Valley, 169 

Cockcrmouth, 276 

Cold Ashby, 210 

Coldest months on record, 

Old Front, 69, 70, see 

Cold Trccline, 178IT 
Condensation; as Cloud, 34- 

41; as Fog, 26-34; ** 

Rain, 41-45; ft stability; 

Trails, 55 
Consett, 205 

Continental 'High', 79, 103 
Convection, 458", 140 
Convergence, 36. 103, 124, 

136, 206 
Cook, James, 7 
Corn Laws, 246 

Cornwall. 73. 74- 75. 77. ><«, 

108, 116, 122, 181, 220 
Cotswolds, 140, 167, 210, 

256, 259- 264 
Cotton Grass Moor, 191 
Coverdalc, 7, 276 
Clowes, 147 

Crib Goch, N. Wales, 159 
Cromer, 88, 89, 141, 200 
Crossfell, 6, 8, 62, 148, 190, 

210, 212, 232 
Crowberry, 191 
Croydon, 199 
Cumberland, 176, 179, lot, 

Cyclone, see Depression 

Dalton, John, 7, 8, 15, lit, 

Daluuspidal, 192 
Dalwhinnie, 180, 202, 208, 

257, 3°2 _ 

Dark Ages, Climate, 235 

Darling, I". Eraser, 33, 55, 


Dartmoor, 128, 137, 205, 

Mj6, 221, 259 

Dartmouth. 151 

Davis, John, 241 
Death Rate, 284-286 
Dceside, 205 
Dcfford, 168 

Dcloc. 152, 176, 260. 271 

Denbighshire, 260 
Denmark, 226 
Dent, 192 

Depressions, U3ff; Rainy Sec- 
tor, 64; Fronts, 66. 60: 
Tracks of, 66, 67; m Cir- 
culation ; Winter Weather ; 
Slimmer Weather 
Derbyshire, 34, 94 
Derham, William, 8 
Derwenrwater, 161, 240 
Devon, 155, 181, 221 ; South, 

112, 116, 151, 284 
Dew, 18, 26, 28, 45, 126; 

point, 28, 30 
Dight, F. H. G., 92 
Dines Anemograph, 16 
Disease, 285 
Divergence, 137, 138 

DO, G. M. B., 53, 55 
Dorset Coast, 114 
Douglas, Ann. 264, 271 
Douglas, C. K. M., 122, 133, 

Dover, 170 
Dover, John, 12 
Drainage of Fields, 246, 296 
Drainage and Habitabilily 

in Moorland Areas, 193 
Drizzle, 41 
Droitwich, 167 
Dry Adiabalic Lapse Rate, 

46, 47, 48. 49 
Dry Spells, 258, 259 
Dry Periods, 245, 246, 247 
Dublin, 130, 270 
Dumas, 3 

Dumfriesshire Valleys, 192 
Dunbar, W., 92 
Dun Fell, 187, 188, 189, 30a 
Dungcncss, 151, •-•:','• 
Dunstable Downs, 140, 150 
Durdham Down, 287 
Durham, 5, 18, 21, 30, 78, 
96, 130, 150, 165, 166, 
168, 187, 255, 259, 269 
Durst, C. S., 55 
Dust Haze, see Haze 

Last Auglia, 129, 153. 169. 

>7', 235, 263, 280 
Eastbourne, 157, 158, 203 
Eddington, Sir A- 7 



Eden Valley, 61, 148, 209 
Edinburgh, 8, ai, 94, 115. 

119, 147, 223, 268 
Egypt, 230 

Elizabethan Weather. 240 
Ellis, Havelock, 275, 297 
F.lrnsione, 170, 268 
English Channel. 67, 75, 85. 

Il6. 2ig 
Ennerdalc, 180 
Equatorial Air, sa Air Masse 
Eskdale, 150 

Eskdalemuir, 17, 195, 270 
Essex, jjt, 116 
Esthwaite, 15 

Europe, Central. 68, 79, 81, 
85, '°7. 231; N.W., 90. 

107; East, 107 
Evaporation. 28, 41, 47 
Exeter, 140, 199 
F.xmoor, 192, 272 
Extremes ol 'I cmperaiurc, 

Faeroe islands. 286 

Falmouth Observatory, 169, 
194, 270 

I'amborough, 157, 171 

February, 103 

Fell Wind, i6t 

Fcnland, 26, 62, 153, 154, 

Iklra, 139 

Fiennes. Celia, 143, 287 

Fife, 137 

lindhorn, 124 

Fisher, James, 289 

Fjords, 162 

I'lamborough Head, 152 

Fleetwood, 262 

Fleure, H.J. 297 

Floods, 264, 265, 266: X4 
Fourteenth Century Flood- 
ing, Chelsea Floods. Moray 
Floods, Great Storm of 


log, 26 34, 74,85, 115, 116, 
■ 751 Radiation, 29, 32, 33, 
I2g; Advection, 31, 32, 33; 
Sea, 32; Horizontal Mix- 
ing, 32, 36 

lohn, 130, 257, 288 

Forres, 255, 288 

Forth, Firth ol, 146 

Fortrose, 119, 160 

Fort William, 191. [95 

Fourteenth Ceniury Flood- 
ing. 237. 238 

Fowey, 169 

France, Northern, 73. 75, 78 
121. !22, 223; East, 106 

Franklin, Benjamin. 198 

Eraser Darling, 145 

Friction, 61, 62. 262 

Frontal Thunderstorms. 122 

Frost, in, 129. 1 6711 ; and 
Altitude, i86ff; Early 
morning, 127; Feathers. 

S,; Ground, 18, 19, 20; 
oar. 26, 45, 79; Hollows, 
164, 165, 169. 170, 172, 
255; in built up areas. 173, 
174; R. Thames, 241. 271 
Fruit Growing Areas. 168. 

Gales, Frequency ol, 120. 

IS81 261; Appendix 
Galloway, 181, 190, 278 
Garuett, A., 193, 196 
Garrigill. 175, iru 
Garsdalc, stc Aisgill 
Germany. 79, 81, no. 11. 
Geochronology. 226 
Gibraltar, 149 
Glaciation, Effects ol, 213- 

218, 224 
Glaciers. 224, 226. 239, 240. 

Glaishcr. 10, 

Glasgow, 29, 85, 142 
Glaslyn Valley, 136 
Glasspoole, J., 13, 23, 94, 

237, 246. 249, 266, 271 
Glenlivct, 151 (sa also Braes; 
Glengarry, 137, 190 
Glenquoich, 258 
Gloucester, 161 sa also 168 
(jodwin, H., 249 
Gold, E.. 82. g'2. 152, 176. 

Gordon Castle, 2t, 260 
Gough. John, 13, 15, 16 
Grainy Ghyll. 179 
Grampians, 138, 150 
Grasmerc, 210 
Great End Gullies. Borrow- 

dale, 210 
Great Lakes, 33 
Great Plains. N. America, 70 
'Great Storm', November 

'7t>3. 237 
Great Wakering. 269 
Greeks. \. 2 

Greenland, 54, 59, G3, 67, 71, 
77, 84, 107, 108. 117. 214 

Greenock, 194 

Greenwich, it. 21; Obser- 
vatory, 21 

Ground Haze, 140 

Growing Season, i8off 

Growth Rings of Trees, 231 
238. 239 

Gudbrandsdai, 34 

'Haar', 115 

Hail, 41, 42, 43, 78, 95, 2' 

262; in U.S.A.. 50 
Hakluyt, 241 
Hciford River, 151 
Halifax, 260 
Hambleton Hills, 264 
Hampstead, 20, 291 
Hancock, D. S., 269, 272 
Handisyde, C. C, 23, 297 
Harding, ).. 122. 133 
Hare, F. K., 55 
Harrogate, 287 
Harrow, 256 
Hartlepool, 156, 157 
Harwood (Teesdalc), 191 
Hawke, E. I... 91, 92, 165. 

172. 176. 166, 272 
Haze, 26, 11." 

Heal Stroke, Limits ol, 292 
Heather, 191 
Healing, 197, 198 
Heavies! Snowfall Areas, 20=, 
Heberden. 13, 16 
Hebrides, 54, 58, 87, 10H, 
'23. '94. 202, 218. 221. 
252, 289 
Heddon, V., 249 
Helm Wind, 62, 148, 149, 
187; similar phenomena. 
150; sa Cross Fell 
Hclvellyn, 210, 211 
I lerdwick Sheep. 186 
Hertfordshire, 50 
Highs, see Anticyclone 
Hoar Frost. ,a Frost 
Holderness, 78 
Holland, 29, 79, 80 
Holmfirth, ja\ 
Holt, 10 
Holyhead, 151 
Hooke. Robert, 7, 1 1 
Hooker, R. H., 237. 249, 297 
Houghall Meteorological 
Station, 165, 166, 168. 257 
Housing, 280, 282 
Hoy. 1 6 

Howard, Luke, 16, 55 

Hoylake, 158 

Huddcrsficld, 205, 304 

Hudlesion, F., 272 

Hull. 88, 158 

Hunter, John, 12 

Huntingdon, Ellsworth, 3, 

liunungdonshirc, 131 

Hutchinson, 10, t6 

Hutton, 13; Theory ol Rain- 
fall, 14, 15 

Huxham, John, 8 

Huxley, 7 

lec Age, 213-17 passim, 239; 
Causes of, 217, 218, 240; 
Climate, 218-223; Inter- 
glacial Periods, 216; Post- 
glacial Changes, 225 
Ice Crystals, 37, 51 
Iceland, 59, 208, 239, 247 
Icelandic Depressions, 24, 66, 

67, 83, 108, 118, 231 
Icelandic Saga, 236 
Icing on Aircraft. 4^ 
Industry and Climate, 282 
Ingleborough, 198 
Instability, 45 ff, 77, 78, no, 

140, 199, 200 
Instrumental Records. 251!! 
Interglacial Periods, 216" 
Inverness, 137, i 59 , 190 
Inversion, 31. 50, 53. 135, 

'Inversion Layer', 82, 129, 

144. '47 
Ireland, 67, 77, 104, 123, 

200, 203, 232, 270, 280 
Irish Gold Ornaments, 2;i 
Irish Sea. 218 
Iron Age. 230, 233 
Ironbridge, 216 
Isle of Man, 155. 272 
Isobars, 62 

Jenyns, Leonard, 12. 18 
July Weather, i2ofl, 210 
June Weather, 1 17IT 
Jurin, 10 

'Katabatics', 162, 1G4 Alaska, 248 

Kcais, 211 

Keele, too 

Kendal, 8, 13. 252, 208. 

Kendrew, W. G., 23, 54 


Kent, 44, 78, 80, 88, 102. 
1 10, 200, 205, 237, 259, 262 
Keswick, 161 
Kew, 11, 17, 21, 23 
Killhopc Summit, 209 
Kilmarnock, 91, 2^5 
Kimble, G. T., 55 
Kindcrscout Peat-Hags, 208 
Kinlochlcven, 194 

Labrador Currenl, 31, 107 

Labrijn, A., 2.(9 

Lady Cross. 261 

Lairig Ghru, 17.^ 

Lake District. 34, 50, 52, 78, 
79. '«6, 136, 142. 190, 
192. 203, 210, 278 

Lakes, effect on winds, 161 

Laki, 247 

Lamb, H. H., 91, 92, 176 

'Lambing Storms', no 

lambing time, 181 

Ijimmcrmuirs, 200, 202, 261 

Lancashire, 21, 33, 43, 61. 
70, 104, 141, 191,238,243 

Langdale, 136, 190, 193 

1-ipse Rnie sa Dry Adiabant 
Lapse Rale 

l.axd.rla. 236 

Leaching, 378 

Leeds. 9, 61 

l-eicesier. 100. 144, 145, i 55 . 

Lcilh, 255 

'Levanter', 149 

Leven, KB 

Lewis, Hebrides, 43 

l.cwis, L. F., 270, 272 

l^wis, W. V., 151, 176, 249. 

, .279, 297 

l.iberton. 255 

I .ibya. 26 

lightning, 44, 70, 77; in 

U.S.A, 50 
Lmdsey, 78 
Lincolnshire, 115-6; Wolds, 

I -indisfarne, 23b 
Little Eaton, 79 
Little I-angdale, 34 
Little Rissington, 168. 256 
Liverpool, 10, 266 
Llangollen, 216 
Lleyn, 155 

Local Variation in Tem- 
perature, i63fT 
Loch I ..i'.mii. 278 
Loch Linn he. 137 


Loch Tay, 208 

Lochnagar, 145 

Loch Ness Monster, 150 

I<ockc, John, 7 

'I.ocss' Soils, 222 

London, 77, 80, 120, 121. 

123, 140, 160, 172, 254 
London Bridge, Old, effect 

on river, 238 
London Fog, ' 5 6, i29,j«For 
Lonk Sheep, 186 
Lossiemouth, 130, 159 
Louth, 262 
l-othians, 4 
I 141 
Ludlam, F. IL, 39, 55 
l.unds, 34 
Lympne, 80 

Macrocarpa, 163 
Madeira, g6 
Maiden Castle, 230, 231 
Maidstone, 257 
Mallerstang, 192 
Malvern, 167, 168; Hills, 127 
Mara Soul, 210 
Manchester, 2g, 75, 143, 144. 

'94. 255 
Manley, G., 118. 133, 177, 

196, aia, 250 
Manncrfell, 239 
Mares' Tails, 38 
Margary, I. D., 23 
Margate, 151, 158. 170 
Maritime Polar Air, 77-78 
Marlborough Downs. 222 
Marlry Common, Surrey, 98 
Marshall, W. A. L., 263, 272 
Marsham Phenological 

Records, 7 
May Snow, 209 
Mayfield, 270 
Mediterranean Climate. 1,2; 

Fruits, 283 
Mcndip Hills, 264 
Meredith, George, 105 
Merle, W., 23 
Merrick Hills, 190 
'Merry Dancers', 270 
Mersey, 85 

Mcsoiilhic Climate. 230 
Mesopotamia, 230 
Meteorological Office. 12, 

13, 20, 90; Record, 21 
Meyer, G. M., 250 
Midlands, 33, 74, 99, 100, 
no, 120, 180, 225; Easi 
29. 62, 260, 263; West, 42 


Migrations of People and 

Ciimate, 233 
Mildcnhnll, 258, 262 
Mill, 11. K.. 12. 23 
Miller, A. A.. 54, 264, 272 
Miller, J. F.. 15, 16, 188, 

Milton, 24 

Minnacrt, M., 146, 177 
Mirrlecs, S. T. A., 92 
Mist. 2G-34, 74. ia 6, ,a 9. 

Moli n 1:1, 139 
Moor house, Westmorland, 

179, 181, 183-5 passim. 

187, 207, 300 
Morcton-in-ihe-Marsh, 167 
Moray. 78, 130, 138, 139 
Moray. Floods of 1829, 124. 

Monxambc Bay, 130,203 
Mountain Breezes, 156 
Mountains, effect on 

wcadier, 137, 138 
Mountains and Moorlands, 233 
Mu-rovy Company, 240 

Nairn, 78, 139, 159. ' 6o - a8a 
Negrctti and Zambra. 10 
Nenlhead, 181 
Neolithic Age. 229, 230 
NewcasUe-on-Tync, 206 
Newfoundland. 31, 66 
New Zealand, 296 
Nicohcm, Bishop, 261 
Night Frost, 102; «« Frost 
Nivation, 222 
Nordcn, 7 
Norfolk, 7, 78, 102, 151,202. 

Norfolk Broads, 286 
Norman Cotii|iie-.t. Climate, 

Norniand, Sir C, 52, 55, 297 
North Africa, 75, 76 
North America, 106 
Northern Lights, 270 
North Pole, 214 
North Sea, 70, 81, 1 13. 1 14. 

119, 121, 122, 200, 218, 
227, 272; Coasts, 79, 
B; Cloud. 82, 87, 98 
Northamptonshire, 44, 78 
Northumberland, 151, 159, 

a 59 
Norway, 34, 59, 66, 67, 70, 

84, 180, 224, a 39 
Norwich, 141, 158 


Nottingham, SI, 263 

Noios, 2 

November Weather, 129(1 

Cakes, 209 

Oats, 280 

Occlusion, 68, 69. see Depres- 

October Weather, 128H 

Oklahoma, 26 

Oldham, 205 

Olive, 163, 283 

Orkney, 21, 2G2, 271,272 

Opalescence, 208 

Orograpliic Rain 122; >et 

Ousby, 8 

Ovey, C. D., 250 

Oxford, 75, 180, 238 

Pain, 10 
Paisley, 255 

Paton, James, 146, 177, 212, 

Peacehaven, 150 
Peach, 163 
Peak District, 150 
Pearsall, W. H., 196. 233, 

Pease, W., 7 

Peat, 191, 192, 226, 223, 232 
Peebles-shire, 169 
Pembroke, 73, 155, 181 
Pembroke, Lady Ann, 2G1 
Pendcnnis Head, 262 
Pendle Hill, 192, 195 
Penistonc, 261 
Penman, H. L.. 279 
Pennant, 211 
Pcnnincs, 7, 61, 136, 140, 

148. 175- 203, 205. 212. 

Pennington, \S'., 220, 250 
Penrith, 112, 155, 161, 183-5 

passim, 256, 257 
Penzance, 21, 1 58 
1'cpys' Diary, 242 
Pcrdiswell, 167, 168 
Perthshire, 20, 183 
Peterborough, 257 
Pctlcrssen, 30, 55 
Phillips, J., 7 
Pine, 226 

Piinlimon Moorlands, J36 
Pliocene, 215 
Plymouth, 8, 137, '9 H 
Polar Air, Continental, 79, 

81. 198; Maritime. 82. 95, 

Polar Air, {Conld.) 

120, 122, 128. 131, 135, 

145. '98 
Polar Sea, 2 1 4 
Pollen Analysis. 226, 230 
Pomerania, 235 
Pontresina, 34 
Portugal, 220 
Poultcr, R. ML, 56 
Precipitation, m Rainfall : 

Pressure, yf/ff; Gradient, 62 
Prestatyn, 130 
Prestwick, 123, 203, 255 
Prestwood. 168 
Primitive Man, 225 
Princctown, 128, 137, 183, 

185, 192 

Qucensbury, 205 

Radbournc, 191 

Radclifl'c Observatory. 21, 

Radiation, 19, 28, 30; ai 
night, 45, 126, 127, 162: 
day, 57, 115; dilfuscd, 193 

Rainfall, 41-45; Ixical Varia- 
tions, 136IV: and Altitude, 
igoff; Range, 2s8ff; 
Heavy, 264, 265; Monthly, 
268; Annual, 279, 280 

Kaisirick. A., 177 

Rarasgate, 158 

Range ofDaily Tempcraturr, 
163, 164 

Reading, 258 

Reafforestation, 180 

Reclamation 01 Moor, 191-2 

Red Pike, Buuermcre, 175 

Red Tarn, 210, 211 

Regent's Park, 175 

Relative Humidity, 18, 25, 
36, 120, 135, 291 

Relaxing Air, 287 

Reykjavik, 178 

Reynolds, R., 272 

Rhyl, 257 

Ribblchcad, 61 

Richmond, Yorkshire, 259 

Kickmansworth, 165, 172. 

Rime, 45 

Rising Air Bubbles, 50 

Robinson, T., 8, 16 

Rochdale, 260 

Roman 1 into, Climate, 229 

Rome, 2 

Ross-shire, 130 

Rotation of the Earth, effects 
of. 59. 60, 61 

Rotliiemurchus Forest, 1 79 

Roxburgh, 4 

Royal Horticultural Society. 

Royal Meteorological 
Society, 9, 12, 21; Quar- 
terly Journal, 22. 53. 91. 

'Running Means', 242, 243 

Russia. 72, 79, 81 

Rutland, 8 

St. Ann's Head, 262 

St. Ives, 158 

St. Luke's Summer, 91 

Salcombe, 151 

Salisbury, Sir E. J., 179, 19G 

Salt Spray, effect on 

tation, 178 
Salzburg, 69 
Saturation, w Atmosphere, 

properties of 
Saxton, 7 
Scandinavia, 79. 81, 99, in. 

214,215, 216 
Scarborough, 156, 158, 170 
Schove, D. J., 250 
Schumann, T. E. W., 56 
Scilly Isles, 198, 202, 262 
Scorer, R. S., 150, 177 
Scorcsby, W., 7 
Scotch Corner, 260 
Scotland, 78, 87, 94, 95, 10G, 
J 18; West, 43, 67, 71, 116. 
121, 130, 255, 260; Easi, 
80, no, 115, 179, mm, 
223; Northern, 98; High- 
lands, 31, 33, 78, 85, 103, 
109, 178 
Scottish Meteorological 

Society, 188 
Scottish Sea Lochs, 162 
Sea Breezes, 156IT 
Sea Ice, 215, 234 
'Sea fret', 1 15 
'Sea rokc', 115 
Sea Temperature, 70 
Seasonal Abnormalities ol 

weather, noff 
Scadiwaitc, 16, 179, 202, 

Seaton, 169 
Sedgwick, A., 7 
Seine. 236 


Selborne, 145 

September, 1250 

Severe winters, 244 

Shakespeare, 4, 252, 274 

Shap Fell, 62. 107 

Shaw, Sir Napier, 92 

Sheffield, 205 

Shelley, P. B., 6, 94, 164, 

Sheppard, P. A., 56 

Sheringham, 200 

Shedands, 179, 262. 270.272 

Short, T., 115, 118 

Shotover Hill, 291 

Shrewsbury, 140 

Shropshire, 37, 106. 107, 136 

Sibelius, a, 93 

Siberia, 79: High, 59, 81, 83, 
107; Winter, 215 

Sid mouth. 169 

Simpson, Sir G., 56, 217, 

Sirocco, 2 

Six, James, 10 

Skiddaw, 142 

Ski-ing in Britain, 202, 208 

Skyc, 120, 140, 146, 209, 

Sleet, 41, 260 

Smoke, effects of, 142, 143, 
'44. «62 

Smith, G., 143 

Smith, K. M., 56 

Smith, W., 7 

Snow, 41, 42, 78, no. i! 1. 
'95. 197-212; and Radi- 
ation, 162, 167; as a con- 
ductor, 171; permanent, 
209-212, 240; Very heavy 
falls, 259; Line, 190 
Snowdon, 130, 190, 192, 212 
Soils as Conductor, 1 7 1 
Solar Radiation, 17, 215, 217 
Solent, 62, 116, 147 
Somerset, 7, igg. 27G 
S. African Veldt, .(.( 
South Coast, 120 
South Georgia, 220 
South Orkneys, 220 
South Uist, 288 
South Wales, 79 
Southport, 136, 157, 159, 

'95. a r>5. 26a, 288 
Soutra Hill, 261 
Speed, J., 6 
Spcnce, ML T., 177 
Speyside. 124, 205, 28H 
Spitsbergen. I] 


Spring Weather, loGfi 
Spurn Head, 200 
Stability, 45-55 
Staininore, 195 
Staffordshire, 263 
Stamp, L. D., 213. 216, 227 

Steers, J. A., 238. 250 
Stephenson, G., 7 
Stevenson, T., 11; Screen, 

11, 19 
Stockholm, 283 
Sloke-on-Treni, M Keele 
Stone, R. G., 297 
Stone Walls, 155 
Stonehcngc, 230 
Stonehnusc Meteorological 

Station, 168 
Stonyhurst, 255. 263 
Stomoway, 123, 151, 194. 

Stow, J., 240, 250 
Straits of Dover, 227 
Siratocumuhis, tigjir* 

Stratosphere, 53 
Sub-Arctic Britain, 188H 
Suffolk, 44, 262 
Summer. 117. 118, i2off 
Summer Kainlhll, tflO 
Sunderland, 151 
Supercooling, 37-40 
Sussex, 42, 44, 77, 88, 102, 

237. 259 
Sutherland. 78, 161, 190. 

Swaffham Hulbcck, 12 
Swalcdale Sheep, 1II6 
Swansea, 255 
Sweden, 216 
Swinbank, W. C, 30, 56 
Symons. G., 12 
Synoptic Weather Map*, 11. 

12, 20 

Tacitus, 29, 235 

!ey, A. G., 23, 196 

Tay, Firth of, 262 

Taylor, G. I., 5G 

Tcesdale, 5, 151, 175, 191, 
192, 209, 260 

Tcignmouth, 183 
lelscombe, 150 

Tcmpcratui- . Diur- 

nal Rang 

tretnes, 251-258; Gradient. 

Ierlinry Era, 214, 215, 216 


Thames Valley, 85, tail 
river frozen, 241, 271 

Think, 263 

Thomson, J., 134 

Thorcsby, R., 9, 23, 143, 

Thunder, 44, 70, 77. "2°. 
121, 122, 195, 262, 263; 
Clouds, 43 

Thunderstorm Census Or- 
ganisation, 263 

Tidal Surges, 237 

Ticrra del ruegO, 220. 296 

Tinn, A. 13., 263, 272 

Tiree, 87, 262 

Todmordcn, 264 

Tomintoul, 202, 279 

Tornadoes, 263 

Torquay, 203 

Totland Bay, 90 

Towneley, R., 7, 9, 23 

Tramontana, 2 

Trccline, set Cold Trceline 

Trent Valley, 262, 263 

Tropical Air, 72-77, 120, 

Troposphere, 53 

Tunbridge Wells, 258 

Tundra, 225 

Turbulence, 26, 32, 35, 39, 

44.73. "9 
Turf Banks, 155 
Tweedside, 128 

Tyndalc, 7 

Tynemouth, 5, 100, 149, 200 

Tyneside, 129 

Tyrol, 69 

Ullswatcr, 161, 179 
United Slates, 49 
Upminstcr, 8 
Upper Air Soundings, 51 


Upper Limit of Settlement, 

Ushaw College, 165, 166, 

168, 187 

Vancouver Island, 296 
Valley Inversion, 126 
Variations of Climate 2i3ff 
Vatnajokuli, 240 
Vegetation and Air Tem- 
perature, 172, 173 
Vertical Ascent of Air, limit 

of. 53 
Vine, 283 
Visibility, 26, 29, 33, 74, 118, 

139, 141-146; see fog 
Volcanic Eruptions, effects 

of, 217-218, 247-248 

Waddinglon, 258 
Wadsworih, J., 279, 297 
Wakefield, 88. 2<-,6 
Wales, 36, 61, 264; N.W. 7 b, 

116, 130, 131, 200; South, 

Walker, Sir C, 51, 170, 

Walled Gardens, 175 
Warmest Months on Record, 

2 54 
Water Vapour, set Cloud etc. 
Watiing Street, 140 
Weald, 169 
Wear, River, 166 
Weardale, 182, 192, 209 
Weather Report, Monthly, 

21, 171, 269; Dailv, 20 
Wells, C, 18 
Wcnlock Edge, 106 
West Kirby, 158 
West Linton, 119 
Westmorland, 223 

Wharfedale, 144, 203 

Wheat, 280 

Wheeler, R. E. Mortimer, 
231, 250 

Whipsnadc, 185 

Whitby, 156, 236 

White, G., 7. 'O, 145. '77. 
247, 250 

Whitehaven, 15 

Whidcy Bay, 158 

Whitworth, .)., 7 

Wick, 102 

Wlgan, 159 

Wind, see Circulation; and 
Settlement, 150; and Vege- 
tation, 151, 153, 154, 179; 
and friction, 262; Rose, 

Windermere, Lake, 226 
Winter Weather, 95II 
Woldinghain, 199 
Woodforde ('Parson'), 247 
Woodland, Clearing, 246. 

Woolacombc, 140 
Worcester, 44, 167 
Wordsworth, 52, 105, 120, 

140, 143, 179 
VVraggc, C, 12 
Wren 1 ham, 7 
Wrexham, 130, 257, 2(i(> 
Wright, W. M., 250 
Wrynose, 34 
WyclifTe, 7 

Yarmouth. 152, 158 

York, 140 

Yorkshire, 33, 115, 136, 202 

273; East, 44; Wolds 205, 


Zcunei, V. E., 250 


1. BUTTERFLIES — E. B. Ford 


— Brian Vesey-FitzGerald 

3. London's natural history 
— R. S. R. Fitter 

4. Britain's structure and 
scenery — L. Dudley Stamp 

5. wild flowers — John Gilmour 
& Max Wallers 

6. natural history in the 
highlands and islands 
— F. Frater Darling 

7. mushrooms and toadstools 
— John Ramsbotlom 


— A. D. Imms 


— W. B. Turrill 


— W. H. Pearsall 

12. THE SEA SHORE — C. M. Yongl 


ILLUSTRATION — Wilfred Blunt 


— T. T. Macan & E. B. Wortltingbm 

AND LIMESTONE — J. E. Lousley 

17. BIRDS AND MEN — E. M. NkhoUoll 

MAN IN BRITAIN — H. J. Fleuie 


— V. S. Summerhayu 


and reptiles — Malcolm Smith 


— L. Harrison Matthews 


— Gordon Manley 


—J. R. Harris 


— Ian Hepburn 

25. the sea coast — J. A. Steers 

a6. the weald — S. W. Wooldridge ^ 
Frederick Goldring 

27. dartmoor — L. A. Harvey & 
D. St. Leger-Gordon 

28. sea-birds — James Fisher & 
R. M. Lackley 

29. the world of the honeybee 
—Colin G. Butler 

30. moths — E. B. Ford 


— L. Dudley Stamp 


— H. L. Edlin 


— John Raven & Max Walters 

the open sea: The World 
of Plankton — Sir Alister Hr.rdy 


—Sir E. John Russell 


- . B. Williams 





37. the open sea: Fish and Fisheries 
— Sir Alister Hardy 


— W. S. Brislowi 


— Edward A. Armstrong 

40. bumblebees — John B. Free Gf 
Colin G. Butler 

41. DRAGONFLIES — Philip S. Corbel, 
Cynthia Longfield & JV. W. Moore 

42. fossils — H. H. Swinnerlon 


— Sir Edward Salisbury 


—K. C. Edwards 


and wales — W. G. Hoskins & 
L. Dudley Stamp 

















Ernest JVeal 

Jo ft n Buxton 

Edward .1. Armstrong 

Stuart Smith 

D. Methersole-Thompson 

M. Rothschild and T. Clay 

JViko Tinbergen 

Frank A. Lowe 

Monica Shorten 

H. V. Thompson & A. j\\ Worden 

Guy Mountfort 

J. W. Jones 

C. T. Prime 

C. M. Tonge 

J. D. Summers-Smith 



"Dr. Ford combines the ardour of the naturalist with the rigorous discipline of the 
experienced research worker. These qualities arc reflected in the book, so that while 
one outstanding feature is the beautiful colour photographs of live butterflies in the 
wild, another is the frequent reference to gaps in our knowledge, which naturalists 
might fill by careful observation. The book is thus not just another butterfly book, 
but is an introduction to the scientific study of butterflies. Evolution is taken as the 
key-note, as we might expect from Dr. Ford, who is himself a geneticist of distinction. 
This is certainly a book that every naturalist, budding or fully fledged, will want to 





"Can be classed unhesitatingly as of the highest value. What is specially to be 
admired in the present volume, over and above the fascination of the subject itself, 
and the obvious pleasure with which the author shares his learning with his readers, 
is the lucid and logical way in which it is arranged." 


"I found this one of the most absorbing books that I have ever come across. It? 
object is to trace through millions of years the geography of the British Isles ami 
so to present a general view of the stage and setting of our Natural History." 


"He writes with such knowledge and illumination. The book is a classic of its 

ki " d -" PUNU. 



"Mr. Nicholson has earned our gratitude anil praise for the masterly way in which 
he has treated liis subject. The value of this book is greatly enhanced by the many 
excellent illustrations." field-marshal viscount alanhrooke 

"A most valuable contribution to British ornithological writing. It is full of personal 
observation and is very readable, but it has also the value of a work of reference even 
the author disdains any attempt (o make it so." 



"This is no ordinary book on British trees and British forestry. It is one which we 
confidently recommend to all who take an interest in trees or to anyone who wishes 
to develop one. The illustrations arc an outstanding feature of the book." 


"The author knows his subject well, and this book is a general survey packed with 
information that is clearly stated. The historical section is good. Mr. Edlin shows 
himself more considerate of asthctic value than most foresters arc. His l»ook can be 
warmly recommended." the times 

'This addition to the New Naturalist series 
maintains the high standard of its pre- 
decessors and should be in the hands of 
every naturalist. It is not a meteorological 
textbook but presents in a clear manner all 
the varied factors which determine the 
circulation of the atmosphere over our 
islands through the four seasons. Professor 
Manley is to be congratulated on provid- 
ing a volume which should give pleasure 
and intellectual entertainment to all who 
love the countryside.' The Naturalist 

'Readers of this learned and delightful 
book will agree that Professor Manley has 
succeeded in his difficult task as no one 
else could have done. It is a book to have 
at one's elbow all the year round, and to 
dip into for more and more enlightenment 
regarding the endless vagaries of our 
climate.' Countryside 

' Professor Manley has assembled his great 
fund of information simply, clearly and 
enticingly. Be the reader farmer or gar- 
dener, rambler or artist he will find here 
a wealth of carefully selected and accurate 
information full of practical interest and 
aesthetic satisfaction. 

' Even if he is none of these, he will 
find enjoyment in the book itself, for it 
contains 73 photographic plates, half of 
them beautifully coloured full-page illus- 
trations of the British scene through the 
season.' Spectator 

'An example of science without tears, and 
many of the illustrations are both instruc- 
tive and beautiful.' 

The Times Literary Supplement 

'This is a book which will give pleasure, as 
well as instruction, to all its readers. 
Their pleasure will be immeasurably 
increased by the 41 magnificent colour 
photographs.' Weekly Scotsman 

Jacket design by 
Clifford and Rosemary Ellis 

The New Naturalist 

■ ■ 


—Brian Vesey-FitzGerald 

4. Britain's structure and 
scenery — Sir Dudley Stamp 

5. wild flowers — John Gilmour 
& Max Walters 


F. Fraser Darling & J. Morton Boyd 

7. mushrooms and toadstools 
— John Ramsboltom 


— A. D. Imms 


— W. B. TurriU 


— W. H. PearsaU 

13. THE SEA SHORE — C. M. Tongi 


illustration — Wilfrid Blunt 


— T. T. Macon & E. B. Worthington 

AND LIMESTONE — J. E. Lousley 

17. birds and men — E. M. Nicholson 



— V. S. Summerhayts 


and reptiles — Malcolm Smith 


— L. Harrison Matthews 


— Gordon Manley 
93. AN angler's ENTOMOLOOY 

—J. R. Harris 


— Ian Hepburn 

35. the sea coast- — J. A. Steers 

a6. the weald— S. W. Wooldridgt & 
Frederick Goldring 

37. dartmoor— L. A. Harvey & 
D. St. Leger-Gordon 

28. sea-birds — Janus Fisher & 
R. M. Lockley 


—Colin G. Butler 

30. moths — E. B. Ford 


— Sir Dudley Stamp 


— H. L. Edlin 


—John Raven & Max Wallers 

34. the open sea: The World 
of Plankton — Sir Mister Hardy 


— Sir E. John Russell 


— C. B. Williams 

37. the open sea: Fish and Fisheries 
— Sir Alister Hardy 


— W. S. Bristowe 


— Edward A. Armstrong 

40. bumblebees — John B. Free & 
Colin G. Butler 

41. dragonplies — Philip S. Corbet, 
Cynthia Longfield & N. W. Moon 

4a. fossils — H. H. Swinnerton 


— Sir Edward Salisbury 


—K. C. Edwards 


and wales — Sir Dudley Stamp & 
W. G. Hoskins 

46. THE BROADS — E. A. Ellis 


— W. M. Condry 



IN Britain — Sir Dudley Stamp 


— Kenneth Mellanby 


5a. woodland birds — Eric Simms 


and the